Demand Response is considered a key factor in shaping and flattening energy consumption peaks (Peak Shaving). Traditional approach of a simplified "higher prices during peak hours and lower peak hours" has not proven itself. In some cases it only moves the peaks to a different time, and in some other cases, consumers shy away from this approach because they do not see any individual benefit. The traditional approach also does not take into account that the market is now open to competition, the consumer can now move freely between services providers, energy companies today compete with each other mainly through programs they present to their customers.
The uniqueness and innovation in this project is in a way also because it's one of the firsts to consider these three factors:
• the need to design should flatten peaks
• Viability Index
• competition in the market,
these factors help develop a DR strategy that consumers will want to adopt, will benefit energy companies, and affect the reduction of consumption peaks.
To support conceptual and functional innovation, it is necessary to develop suitable technologies and optimization algorithms. These algorithms will need to nalyze the behavior and needs of consumers and generate and analyze projections of use, not only of the collectives but also individual consumers. They need to also examine the behavior and constraints of the network, and the proposals of competitors to optimize proposals to customers. The primary objective of the company under this plan, is to develop a solution that will help energy providers optimize DR programs and offer different programs to different customers while weighing a range of the supplier objectives centering on peaks reduction (Peak Shaving), but also taking into account maximizing profit, attracting new customers and retention of existing customers, using data and forecasts consumption of each client.
A major factor that will affects the success or failure of the DR is non-technical loss, electricity consumed it but not paid for, both as a result of fraud, or failures in processes or systems. Non- technical loss can reach around 10% of the production. Clearly the non-paying customers are not affected by DR. the prevention of non-technical losses is in its infancy in the energy market, the prevailing approach is to perform a lot of very expensive field tests. A second area of research the company's in this project is the optimization of non-technical loss detection (optimization of false-positive and false-negative) that will include algorithms that combine machine learning technologies and rules engines and will help shaving the peaks