Monetization of Advanced Metering Infrastructures (AMI) with Big Data Approach

Indian electricity utility companies have long been under scrutiny for their poor management of energy consumption and assets. Energy thefts, rising green house gas emissions and poor match with demand and supply of energy are some of the struggles of using traditional energy meters. The answer to this point at issue is advanced metering infrastructures (AMI) facilitated with Big Data Analytics which can impart granular insights into customer usage motif.

Smart grids are conducive to superior electricity delivery infrastructures. So, utilities are attempting to pursue high volume data management to harvest dynamic behavioral patterns in the physical grid. Blending and crunching variegated data sets can yield more insights and knowledge that will need to be acted upon. But, analyzing this amassed wealth of information is an extensive trial. Analytics and Big Data are intrinsically linked to each other as application and foundation layers and reinforces the monetization of assets.

There are three domains infused in Smart Grid Utilities Analytics:

Enterprise Analytics- Grid Visualization and Business Intelligence equipped with Predictive Analytics
Grid Operations Analytics- Optimization and Operational Intelligence, Management of assets and outage, Fault Detection and Correction
Consumer Analytics- Behavioral analytics and Load Flow, Data Integration on social media

A convenient advantage of using AMI is that it enables bidirectional communication between the consumer and utilities. From a consumer perspective, apart from management of energy consumption and outage using AMI, reduction of energy bills and greenhouse gas emissions is an ancillary benefit. With more control and power over the amount of energy consumption with the assistance of Big Data Analytics, consumers and utilities alike will be adept in keeping track of energy consumption, load, ROI, revenue losses, energy T&D, service levels and outage, improving SLA ( service-level-agreement). India is expected to install 130 million smart meters by 2021.

Data acquisition is not the challenge. The struggle to keep the management staff informed of their analysis fast enough is the critical problem. The ability to enable operational decisions through transformation of data into actionable intelligence and feeding them to the consumers is not possible if the rate of analysis is low. Execution of data analysis is not pronounced in utilities due to lack of skill to visualize, analyze and comprehend.

If we look into the Indian Smart Grid, the major hurdle is the availability of funds. To tackle this financial challenge, compelling smart grid consumer products, vendor partnerships and a willing investment body seems like the answer. Energy thefts and commercial losses, common in the country unlike developed nations can be evaded by use of tamper-proof smart meters with a prepaid system. But, for operational success, AMI will require faultless connectivity and India is not at par, in this regard, with other developed nations to employ a communication of the extent required for Smart Grids. The capital and hardware costs escalates if investment is done on an excellent communication network setup. This may impact during service deliveries in a negative manner. Due to the lack of awareness, utilities in India need to shift their focus to educating the consumers of the overall potential of Smart Grids rather than mere adoption of smart meters. Now, the benefits from the Smart Grids may be favorable. But, when mismatched policies and incentives come into the picture, the investment seems less attractive. The risk factor associated with utilizing Smart Grids is shared by every individual in the value chain. Maintaining this, will help align the incentives, tackle policy disputes and offer promising solutions.

If all the above mentioned challenges are catered to, the utilization of Smart Grids and Big Data in the country will not only accentuate sustainable use but also, monetize AMI data. MachinePulse, a Mumbai based IoT platform provider is soon, looking forward to incorporating smart grid data analytics and solutions and transform the smart meter data silos into actionable insights.

About MachinePulse

MachinePulse^TM is an Internet of Things platform provider based in Mumbai. The big data platform, erixis^TM, has in-built machine learning and predictive analytics capabilities. MachinePulse^TM has branch offices in Chennai and Bangalore. Email us to learn more.

SolarPulse 3.0 By MachinePulse Makes Solar Power Projects Smarter

Posted by MachinePulse Tech Pvt. Ltd.

SolarPulse^TM3.0 uses real-time analytics and artificial intelligence to offer performance insights thereby enhancing operational efficiency of solar power plants.

MachinePulse^TM, launches SolarPulse^TM 3.0, an analytics solution for the solar power industry. Catering to industrial, commercial and utility sectors, SolarPulse^TM 3.0 is the new and improved version of its predecessor SolarPulse^TM 2.0, which was introduced in 2013.

The utility solution’s scalable (horizontally and vertically) platform allows monitoring of multiple parameters through no node locking, allowing the user to find the root cause of performance issues easily. All devices are dynamically compared against the best performing device to enable users to spot performance issues through bird eye monitoring on a real-time basis. Users can define their own rules and notifications for benchmarking to a string level granularity. The geographical plant map feature enables quick fault detection through one single dashboard for viewing device performance through a color coded status. SolarPulse^TM 3.0 for rooftop applications is offered through the erixis^TM cloud platform. The cloud solution offers features that equip solar plant owners with the right tools to achieve optimum performance of their power plants. It offers unification of all data on a single platform, all existing as well as upcoming roofs and an interactive platform for multi-roof comparisons and drill down. Through centralized monitoring of rooftops, geo visualizations of all sites as well as individual rooftops are made available.

SolarPulse^TM 3.0 offers stunning, interactive data visualization with a user friendly interface for configuring new devices and parameters for individual needs. Furthermore, SolarPulse^TM3.0 addresses the drawbacks of faulty predictions by using artificial intelligence based models to accurately predict performance, irradiation, maintenance issues, etc. With these predictions, users can make informed decisions based on real-life accurate data.

About MachinePulse

MachinePulse^TM is an Internet of Things platform provider based in Mumbai. The big data platform, erixis^TM, has in-built machine learning and predictive analytics capabilities. MachinePulse has branch offices in Chennai and Bangalore. To learn more, email info@machinepulse.in

Why Factories Using Artificial Intelligence Approaches Are The Future

Posted by MachinePulse Tech Pvt. Ltd.

Artificial Intelligence (AI) can be advantageous, especially where the data exhibits some form of non-linearity.
To reproduce some of the flexibility and power of the human brain, artificial neural networks (ANN) try and emulate the brain structure through computational models.
Different personality types should also be taken into consideration while designing appropriate tools and solutions.

A smart factory with interlinked smart systems is powerful – it responds to customer demands and changes within a short span of time, makes better predictions of future operations, allows for more informed decisions that improve efficiency, etc. However, smart factories are yet to function independently without human involvement. Embodying accurate human-like thinking into a mechanical system will create the most efficient and optimized factories and manufacturing units.

Fuzzy Logic and Neural Network in AI

When it comes to tasks requiring delicate human intuition based decisions, Artificial Intelligence (AI) can be advantageous, especially where the data exhibits some form of non-linearity. Fuzzy logic (FL) and neural network theory (NN) are complementary constituents of AI rather than competitive as it is evident and it is beneficial to employ FL and artificial NN in combination rather than exclusively. To reproduce some of the flexibility and power of the human brain, artificial neural networks (ANN) try and emulate the brain structure through computational models. Suppose, there are five machines running in parallel performing different tasks and one of them gives out an undesired output due to some unseen problem in the machine. Ideally, an intelligent system will retrieve cases from memory that are relevant to solving it and map the solution of that case to the target problem. It will not only map the solution to the problem but, will also adapt the solution as required to fit the new case and will apply the new solution in the real world and store this solution under a new case for future use. Now, in a situation like above, if the other four machines are dependent on the output of the first machine, it might also be possible that these machines might face similar problems which might compromise the safety of the unit. To prevent this, the first system will alert the other systems regarding the faulty situation and the other systems will be able to similarly map out solutions accordingly. This is a classic example of machines adapting and displaying flexibility to different situations independently using a combination of FL and ANN. However, using anecdotal evidence as the main operating principle with no statistically relevant data to support a generalization, is not enough to prove that the generalization is accurate. It is also essential to remember that human beings are not completely rational. Different people might react differently to the same situation and make decisions based on their own unique experiences. Different personality types should also be taken into consideration while designing appropriate tools and solutions.

FL and ANN are not widely used in the manufacturing industry however, once the methods gain traction, and familiarity, the process of configuring such a system will become relatively easy. It is clear that this technological shift in manufacturing industries is here to stay with the promise to improve the quality, safety and precision.

MachinePulse is bringing the groundbreaking results of artificial intelligence approaches to factories through FactoryPulse. The Internet of Things solution has AI capabilities that improve the overall efficiency of projects leading to better performance and higher returns.

About MachinePulse
MachinePulse™ is an Internet of Things platform provider based in Mumbai. The big data platform, erixis™, has in-built machine learning and predictive analytics capabilities. MachinePulse is based in Mumbai, India with branch offices in Chennai and Bangalore. Contact us to learn more.

MachinePulse To Sponsor India M2M + IoT Forum 2015

Posted by MachinePulse Tech Pvt. Ltd.

MachinePulse is a proud sponsor of the India M2M + IoT Forum 2015 that will take place on February 19-20, 2015 at the Royal Plaza, Delhi. MachinePulse will give an industry presentation on February 19 and participate in a panel discussion titled ‘Opportunities & Challenges – WHO HELPS? Real or CREATED?’ on February 19 and another panel discussion titled ‘The Smart City Vision – turning the Smart City Vision into REALITY’ on February 20. To view the entire program, visit the event website.

The India IoT+M2M Forum welcomes reputable brands, professional service providers and distributors in Business, Business Development, Networking and Internet segments. It offers to display a broad range of items and professional services to different industrial verticals.

To schedule a meeting with us at the India M2M + IoT Forum, please email sales@machinepulse.in

MachinePulse Will Be At RE-Invest 2015 In New Delhi On Feb. 15-17

Posted by MachinePulse Tech Pvt. Ltd.

The MachinePulse team will be at the first Renewable Energy Global Investors Meet & Expo (RE-Invest 2015) on February 15-17 in New Delhi. Visit us at stall number 76 to learn about our Internet of Things solutions for the renewable energy industry.

RE-Invest is organized by the Ministry of New and Renewable Energy (MNRE), Government of India as a follow-up to the ‘Make in India’ initiative launched by the Prime Minister of India. The central theme of RE-Invest is to attract large scale investments for the renewable energy sector in India. RE-Invest signals India’s commitment to the development and scaling up of renewable energy to meet its energy requirements. The event is expected be attended by over 200 investors and over 1000 delegates, both domestic and international. Additionally, representatives from the State Government, Public Sector Enterprises, renewable power developers and manufacturers, state renewable energy nodal agencies, and other related stakeholders will play important roles.

To schedule a meeting with us at RE-Invest, please email sales@machinepulse.in

MachinePulse™ Takes Its Obsession With Data to the Factory Shop Floor With FactoryPulse™

Posted by MachinePulse Tech Pvt. Ltd.

MachinePulse™ launches FactoryPulse™ to address the emerging requirement of manufacturing shop floors to achieve operational efficiency and utilize data churning out from devices to provide real-time insights and help make better, more informed decisions.

FactoryPulse™, armored with erixis™ platform’s decision science capabilities, is a solution for machine data generated by various factory equipment, control devices, SCADA systems, sensors, networks, applications and end users.

The solution utilizes the capabilities of MachinePulse™ in the areas of factory monitoring with a deep focus on asset performance diagnostics and performance benchmarking.

Examples of what can be gained through the analytics spectrum covered by FactoryPulse™ are:

Root-cause analysis and remote troubleshooting: Drill down intelligence for key performance indicators ; find root causes behind production downtime.
Performance benchmarking: View best performing machine output characteristics; benchmark against expected performance parameters and set standards.
Portfolio management: Manage an entire plant portfolio with a single tool.
Detect anomalies and outliers: Identify outliers which may be an early warning for issues in device production or deployment.
Flexible integration: Unify data from all devices, API level integration with third party software.

Manufacturing organizations can benefit from improved asset utilization by running preventive maintenance on critical infrastructure equipment and machinery for improving throughput and utilization.

FactoryPulse™ launches with a project concerning overall equipment effectiveness at a leading steel service center in the automotive cluster in Pune, Maharashtra.

MachinePulse Partners With RIIDL Somaiya Vidyavihar

Posted by MachinePulse Tech Pvt. Ltd.

We are pleased to announce that MachinePulse is an official Partner of the Research Innovation Incubation Design Lab (RIIDL) Somaiya Vidyavihar, an innovation centre and accelerator at the Somaiya campus in Mumbai. MachinePulse and RIIDL will be working together to strengthen the culture of innovation and learning at RIIDL by collaborating on students’ projects and events. This partnership furthers MachinePulse’s commitment to building a community of Internet of Things academics, professionals and scholars by fostering innovative and futuristic ventures.

For further information about RIIDL, please visit their website.

About MachinePulse

MachinePulse™ is an Internet of Things platform provider based in Mumbai. The big data platform, erixis™, has in-built machine learning and predictive analytics capabilities. MachinePulse is based in Mumbai, India with branch offices in Chennai and Bangalore. Contact us to learn more.

Machine Learning Techniques For Better And More Efficient Solar Power Plants

Posted by MachinePulse Tech Pvt. Ltd.

Machine learning techniques support better solar power plant forecasting.
Machine learning techniques play a crucial role in deciding where to build a plant when accurate or limited location data is available.
Machine learning techniques help maintain smart grid stability.

The global solar photovoltaic (PV) installed capacity in 2013 was 138.9 GW and it is expected to grow to over 455 GW by 2020. However, solar power plants still have a number of limitations that prevent it from being used on a larger scale. One limitation is that the power generation cannot be fully controlled or planned for in advance since the energy output from solar power plants is variable and prone to fluctuations dependent on the intensity of solar radiation, cloud cover and other factors. Another important limitation is that solar energy is only available during the day and batteries are still not an economically viable storage option making careful management of energy generation necessary. Additionally, as the installed capacity of solar power plants grows and plants are increasingly installed at remote locations where location data is not readily available, it is becoming necessary to determine their optimal sizes, locations and configurations using other methods. Machine learning techniques provide solutions that have been more successful in addressing these challenges than manually developed specialized models.

Accurate forecasts of solar power production are a necessary factor in making the renewable energy technology a cost-effective and viable energy source. Machine learning techniques can correctly forecast solar power plant generation at a better rate than current specialized solar forecasting methods. In a study conducted by Sharma et al, multiple regression techniques including least-square support vector machines (SVM) using multiple kernel functions were used in the comparison with other models to develop a site specific prediction model for solar power generation based on weather parameters. Experimental results showed that the SVM model outperformed the others with up to 27 percent more accuracy.

Furthermore, machine learning techniques play a crucial role in assisting decision making steps regarding the plants location selection and orientation selection as solar panels need to be faced according to solar irradiation to absorb the optimal energy. Conventional methods for sizing PV plants have generally been used for locations where the required weather data (irradiation, temperature, etc.) and other information concerning the site is readily available. However, these methods cannot be used for sizing PV systems in remote areas where the required data are not available, and thus machine learning techniques are needed to be employed for estimation purposes. In a study conducted by Mellit et al., an artificial neural network (ANN) model was developed for estimating sizing parameters of stand-alone PV systems. In this model, the inputs are the latitude and longitude of the site, while the outputs are two hybrid-sizing parameters. In the proposed model, the relative error with respect to actual data does not exceed 6 percent, thus providing accurate predictions. This model has been evaluated on 16 different sites and experimental results indicated that prediction error ranges from 3.75-5.95 percent with respect to the sizing parameters. Additionally, metaheuristic search algorithms address plan location optimization problems by providing improved local searches under the assumption of a geometric pattern for the field.

Lastly, to maintain grid stability, it is necessary to forecast both short term and medium term demand for a power grid with renewable energy sources contributing a considerable proportion of energy supply. The MIRABEL system offers forecasting models which target flexibilities in energy supply and demand, to help manage the production and consumption in the smart grid. The forecasting model combines widely adopted algorithms like SVM and ensemble learners. The forecasting model can also efficiently process new energy measurements to detect changes in the upcoming energy production or consumption. It also employs different models for different time scales in order to better manage the demand and supply depending on the time domain.

Ultimately, machine learning techniques support better operations and management of solar power plants.

How Predictive Analytics Can Make the Renewable Energy Industry Grow

Posted by MachinePulse Tech Pvt. Ltd.

Renewable energy technologies are quickly gaining acceptance globally as a reliable source of electricity. Total global renewable energy installations have increased from 160 GW in 2004 to more than 1,560 GW in 2013. With a growing installed capacity of renewable energy plants comes a growing number of remote monitoring solutions to track the performance of these plants. Enormous amounts of data are being generated by these renewable energy plants and it is becoming ever important to create valuable insights from this data. Big data analytics performed on the data collected from these plants, enables owners and O&M crews to operate the renewable plants at the plants maximum potential. Among all the types of big data analytics that could be performed on the plant data, predictive analytics holds the most promising of providing insights by leveraging performance data to create correlations and outcomes. Let us understand how it could impact on Renewable Energy Industry.

“Data is not Information, Information is not Knowledge and Knowledge is not Wisdom”.

From data to decision science (Source: infogineering)

There are multiple steps required to reach from data collection to generating actionable insights. Predictive analytics is the link in this chain that takes us from the ‘Information’ stage to ‘knowledge’ stage. It models the cause and effect relationship among various parameters using various data mining techniques, statistical models and machine learning techniques thus allowing us a window to see the contextual future events.

According to a study, 20% – 40% of renewable energy cannot be used because it is unstable.

Predictive analytics when used deftly on renewable energy power plants can provide accurate energy production forecasts. It also predicts the machine breakdowns or glitches thereby optimizing overall operational efficiencies. The analytics checks for the correlation of various parameters like irradiation, wind speed, temperature, humidity, cloud cover, transformer status etc. and learns their cause and effect relationship. One study estimates that a good predictive model can increase the power generating capacity of a wind farm by about 10%, which practically revitalizes the entire business. It is also important to note that Predictive Analytics doesn’t only improve operational efficiencies but also improves the life span of the valuable renewable energy technology assets.

Banks are yet to consider renewable energy projects as a sound investment compared to oil and gas power projects.

The current growth of renewable energy technologies could be amplified if there is enough data to prove that they are credible investment options. Numerous renewable energy power projects still lack appropriate funding because of the lack of historic data that raises suspicions on the long term viability of the projects. Predictive analytics can addresses this problem by accurately forecasting energy generation based on historic performance, weather and other parameters. These quantifiable results associated with revenues generated from the future performance can improve the bankability of renewable energy projects.

Ultimately, predictive analytics is set to provide immense value to the renewable energy industry. It is now up to the plants owners to capitalize on this statistical tool to achieve the most out of their renewable energy power plant.

MachinePulse to demo SolarPulse at IIT-Bombay on Nov. 28, 2014

Posted by MachinePulse Tech Pvt. Ltd.

MachinePulse will give a live demonstration of SolarPulse at IIT-B on Nov. 28, 2014

MachinePulse will host a live demonstration of SolarPulse, our solar power plant remote monitoring and control solution, on November 28, 2014 at IIT-Bombay’s campus as part of a short-term course entitled ‘Solar PV Modules and System Testing and Characterization’. The three day course is jointly organized by the Continuing Education Program of IIT Bombay and the National Centre for Photovoltaic Research and Education (NCPRE).

During the period of this course, students will be given the opportunity to visit the 1 megawatt (MW) solar photovoltaic (PV) plant installed at the IIT-Bombay campus. Niketan Shitole, System Integration Engineer at MachinePulse will guide students through the different components of the 1 MW solar power plant. After the plant visit, Niketan will host an exclusive live demo of SolarPulse by logging into IIT-Bombay’s 1 MW solar power plant account. During this live demo, students will view the plant’s live generation and performance data. Niketan will also explain the parameters used to measure a solar power plant’s performance, track past performance and the process followed to respond to plant related alerts and alarms. Students will also be introduced to iceBOX™, MachinePulse’s smart hardware device that collects the data generated from solar power plants, aggregates it and sends it to the server where it is analysed and the end result is presented on the SolarPulse dashboard.

The course begins on November 26, 2014 and ends on November 28, 2014. On completing the course, students can expect to gain an intermediate level understanding of test and characterization techniques, the types of solar photovoltaic (PV) modules, balance of systems components, PV systems and the technical complexities of a solar PV remote monitoring and control software.

To learn more about the event please click here to visit the NCPRE site.