A method and system for early warning of power energy consumption

Abstract

The present invention discloses an electric energy consumption early warning method and system, which belongs to the field of electric energy consumption early warning. The method can firstly carry out a primary electric energy consumption early warning to realize a rough early warning based on predicted power consumption, and then construct a target constraint based on the predicted power consumption, and according to the target constraint, adopt an intelligent optimization algorithm to determine the minimum electric energy consumption, and finally carry out a secondary electric energy consumption early warning according to the minimum electric energy consumption, thereby realizing a more precise early warning and realizing a comprehensive early warning of electric energy, so that the staff can carry out timely dispatch and improve the management efficiency of electric energy.

Description

Technical Field

[0001] The present invention belongs to the field of electric energy consumption early warning, and in particular relates to an electric energy consumption early warning method and system. Background Art

[0002] Electric energy usually refers to the energy generated by power generation, and power generation generally relies on electric coal as the mainstream power generation consumables. Electric coal refers to the coal used for thermal power generation. Electric coal is imported from the coal yard into the coal furnace, burned in the boiler to generate power and converted into electricity, and finally transmitted to the grid for sale. The electric coal needed by power plants is mainly ordered in a centralized manner. Through the annual national coal ordering conference, the state issues coal allocation plans and allocation plans, and then the supply, transportation, and demand links negotiate and adjust them at the ordering conference. With the progress of the market-oriented reform of electricity, the state has also stepped up efforts to promote the market-oriented reform of coal prices. However, coal prices fluctuate frequently, railway and road transportation forces are sometimes tight and sometimes loose, and coupled with the impact of severe weather, power generation companies have repeatedly experienced tight supply of electric coal. Summary of the invention

[0003] The present invention provides an electric energy consumption early warning method and system, which are used to solve the technical problems existing in the prior art.

[0004] In one aspect, the present invention provides a method for early warning of power energy consumption, comprising:

[0005] After learning the historical electricity consumption characteristics using a deep learning algorithm, an electricity consumption prediction rule is obtained, and based on the electricity consumption prediction rule, the electricity consumption in future time periods is predicted to obtain the predicted electricity consumption;

[0006] Based on the predicted power consumption, determining the first-level power energy consumption warning result corresponding to the target power supply area; wherein the first-level power energy consumption warning result includes sufficient power energy or insufficient power energy;

[0007] When the first-level power energy consumption warning result is that the power energy is sufficient, the predicted power consumption is used as a constraint and an intelligent optimization algorithm is used to determine the minimum power energy consumption;

[0008] Based on the minimum amount of electric energy consumption, determine the number of days that electric energy can be consumed, and based on the number of days that electric energy can be consumed, determine the second-level electric energy consumption warning result corresponding to the target power supply area; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0009] When the first-level electric energy consumption warning result or the second-level electric energy consumption warning result is insufficient electric energy, electric energy consumption warning information is generated.

[0010] Furthermore, after using the deep learning algorithm to learn the historical electricity consumption characteristics, the electricity consumption prediction rules are obtained, including:

[0011] Obtain the historical power consumption characteristics corresponding to the target power supply area and the power consumption corresponding to the historical power consumption characteristics;

[0012] Determine the historical electricity consumption characteristics corresponding to M time points as sample data, and use the electricity consumption corresponding to the M+1th time point as label data;

[0013] A deep learning algorithm is used to learn the relationship between sample data and label data to obtain electricity consumption prediction rules.

[0014] Furthermore, based on the power consumption prediction rule, the power consumption in the future period is predicted to obtain the predicted power consumption, including:

[0015] The current power consumption feature of length M is obtained, and the current power consumption feature is identified using the power consumption prediction rule to obtain the predicted power consumption corresponding to the future time period.

[0016] Furthermore, based on the predicted power consumption, a first-level electric energy consumption warning result corresponding to the target power supply area is determined, including:

[0017] Determine the maximum power generation of all generators in the target power supply area in the future period;

[0018] Determine whether the maximum power generation in the future period is greater than the predicted power consumption corresponding to the future period. If so, determine that the first-level power energy consumption warning result is sufficient power energy; otherwise, determine that the first-level power energy consumption warning result is insufficient power energy.

[0019] Furthermore, when the first-level power energy consumption warning result is that the power energy is sufficient, the predicted power consumption is used as a constraint, and an intelligent optimization algorithm is used to determine the minimum power energy consumption, including:

[0020] When the first-level electric energy consumption warning result is that the electric energy is sufficient, construct an objective function for determining the minimum electric energy consumption amount and construct a target constraint based on the predicted power consumption; wherein the objective function is used to characterize the cost function of all generator sets, and the target constraint is used to characterize the operating constraint of the generator set;

[0021] Based on the target constraint, the power generation of all power generating units in the future is initialized to obtain a power generation code, and multiple power generation codes are repeatedly obtained;

[0022] According to the objective function, determine the objective function value corresponding to each power generation code;

[0023] Based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, and the update is repeated until the total number of updates reaches the preset maximum number of updates, and the power generation code with the minimum objective function value is obtained;

[0024] Decode the power generation code with the minimum objective function value to determine the final power generation of each generator set at a future time;

[0025] The pre-stored amount of electric energy consumption required for a unit of power generation is obtained, and the minimum amount of electric energy consumption is determined based on the amount of electric energy consumption required for a unit of power generation and the final power generation of each generator set in the future.

[0026] Furthermore, based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, including:

[0027] Determine a first update ratio and a second update ratio, and divide all power generation codes into a first code group and a second code group according to the first update ratio and the second update ratio;

[0028] The first power generation code in the first coding population is updated by using the best neighborhood exploration strategy to obtain an updated first power generation code;

[0029] A multi-information interaction strategy is adopted to update the second power generation code in the second coding population to obtain an updated second power generation code;

[0030] A global greedy search strategy is used to update the first power generation code after one update and the second power generation code after one update, to obtain the first power generation code and the second power generation code after two updates, and to complete an update process for each power generation code.

[0031] Furthermore, based on the minimum power energy consumption, determining the number of days that the power energy can be consumed includes:

[0032] Determine the current remaining power energy consumption corresponding to the target power supply area and obtain the replenishable power energy consumption corresponding to the future period; wherein the replenishable power energy consumption is obtained from the procurement system of the target power supply area;

[0033] The current remaining power energy consumption corresponding to the target power supply area is added to the replenishable power energy consumption, and then the minimum power energy consumption is subtracted to obtain the comprehensive remaining power energy consumption;

[0034] The number of days that electricity energy can be consumed is obtained by dividing the comprehensive remaining electricity energy consumption by the minimum electricity energy consumption.

[0035] Furthermore, based on the number of days during which electric energy can be consumed, a second-level electric energy consumption warning result corresponding to the target power supply area is determined, including:

[0036] Based on the number of days in which the electric energy can be consumed, determine whether the number of days in which the electric energy can be consumed is greater than a preset time threshold. If so, determine that the second-level electric energy consumption warning result is sufficient electric energy; otherwise, determine that the second-level electric energy consumption warning result is insufficient electric energy.

[0037] Furthermore, after the electric energy consumption warning information is generated, it also includes: transmitting the electric energy consumption warning information to a device or information system designated by a staff member.

[0038] On the other hand, the present invention provides an electric energy consumption warning system, comprising: a prediction module, a first warning module, a consumable quantity determination module, a second warning module and a warning information generation module;

[0039] The prediction module is used to learn historical power consumption characteristics using a deep learning algorithm to obtain power consumption prediction rules, and predict power consumption in future time periods based on the power consumption prediction rules to obtain predicted power consumption;

[0040] The first warning module is used to determine the first-level electric energy consumption warning result corresponding to the target power supply area based on the predicted power consumption; wherein the first-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0041] The consumables quantity determination module is used to determine the minimum power energy consumables quantity by using the predicted power consumption as a constraint and an intelligent optimization algorithm when the first-level power energy consumption warning result is that the power energy is sufficient;

[0042] The second warning module is used to determine the number of days that electric energy can be consumed based on the minimum electric energy consumption, and determine the second-level electric energy consumption warning result corresponding to the target power supply area based on the number of days that electric energy can be consumed; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0043] The warning information generation module is used to generate power energy consumption warning information when the first-level power energy consumption warning result or the second-level power energy consumption warning result is insufficient power energy, and transmit the power energy consumption warning information to a designated device or information system.

[0044] The present invention provides an electric energy consumption early warning method and system, which can first perform a first-level electric energy consumption early warning to achieve a rough early warning based on predicted power consumption, and then build a target constraint based on the predicted power consumption, and according to the target constraint, use an intelligent optimization algorithm to determine the minimum electric energy consumption, and finally perform a second-level electric energy consumption early warning based on the minimum electric energy consumption, thereby achieving a more precise early warning and a comprehensive early warning of electric energy, allowing staff to carry out timely scheduling and improving the management efficiency of electric energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0046] Figure 1 A schematic diagram of a flow chart of an electric energy consumption early warning method provided by an embodiment of the present invention.

[0047] Figure 2 A schematic diagram of the structure of an electric energy consumption early warning system provided by an embodiment of the present invention.

[0048] Among them, 201 is a prediction module, 202 is a first warning module, 203 is a consumable quantity determination module, 204 is a second warning module, and 205 is a warning information generation module.

[0049] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0050] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0051] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0052] like Figure 1 As shown, an embodiment of the present invention provides an electric energy consumption early warning method, comprising:

[0053] S101, after learning historical electricity consumption characteristics using a deep learning algorithm, obtain electricity consumption prediction rules, and predict the electricity consumption in future time periods based on the electricity consumption prediction rules to obtain predicted electricity consumption;

[0054] For example, a BP (Back Propagation) neural network or an LSTM (Long Short-Term Memory) neural network can be used to learn historical electricity consumption characteristics to obtain predicted electricity consumption.

[0055] Since the electricity consumption characteristics used to predict electricity consumption are not complex, the embodiment of the present invention preferably uses a BP neural network to learn the historical electricity consumption characteristics.

[0056] S102, based on the predicted power consumption, determining a first-level electric energy consumption warning result corresponding to the target power supply area; wherein the first-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0057] When the first-level power energy consumption warning result is insufficient power energy, it indicates that the current power generation upper limit no longer meets the power energy consumption of the target power supply area, so a warning can be issued.

[0058] S103, when the first-level electric energy consumption warning result is that electric energy is sufficient, using the predicted power consumption as a constraint, and using an intelligent optimization algorithm to determine the minimum electric energy consumption;

[0059] Although the first-level power energy consumption warning results analyze the power generation capacity, in actual situations, it is very likely that due to irregular changes in transportation and load, there will be a shortage of electric coal, thus failing to achieve the planned power generation. Therefore, the embodiment of the present invention can predict the minimum amount of power energy consumption, so as to analyze whether the power energy consumption is sufficient.

[0060] It is worth noting that the embodiment of the present invention mainly provides early warning for the power energy consumption of thermal power plants, so only the stock of power coal needs to be paid attention to.

[0061] S104, based on the minimum amount of electric energy consumption, determining the number of days that electric energy can be consumed, and based on the number of days that electric energy can be consumed, determining the second-level electric energy consumption warning result corresponding to the target power supply area; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0062] For example, when the number of days that electric energy can be consumed is less than a certain threshold, it can be considered that the inventory of electric energy consumables no longer meets the requirements, and an early warning can be issued.

[0063] S105: When the first-level electric energy consumption warning result or the second-level electric energy consumption warning result is insufficient electric energy, electric energy consumption warning information is generated.

[0064] The present invention provides an electric energy consumption early warning method, which can first perform a first-level electric energy consumption early warning to achieve a rough early warning based on predicted power consumption, and then build a target constraint based on the predicted power consumption, and according to the target constraint, use an intelligent optimization algorithm to determine the minimum electric energy consumption, and finally perform a second-level electric energy consumption early warning based on the minimum electric energy consumption, thereby achieving a more precise early warning and a comprehensive early warning of electric energy, allowing staff to carry out timely scheduling and improving the management efficiency of electric energy.

[0065] In the embodiment of the present invention, after learning the historical electricity consumption characteristics using a deep learning algorithm, the electricity consumption prediction rules are obtained, including:

[0066] Obtain the historical power consumption characteristics corresponding to the target power supply area and the power consumption corresponding to the historical power consumption characteristics;

[0067] Determine the historical electricity consumption characteristics corresponding to M time points as sample data, and use the electricity consumption corresponding to the M+1th time point as label data;

[0068] A deep learning algorithm is used to learn the relationship between sample data and label data to obtain electricity consumption prediction rules.

[0069] The power consumption characteristics may include season data, weather data, temperature data, date type data and corresponding power consumption of the target power supply area at continuous time points, wherein the season data includes spring, summer, autumn or winter, the weather data includes rainy, sunny, cloudy or snowy, and the date type data includes working day or non-working day;

[0070] In an embodiment of the present invention, the power consumption in a future period is predicted based on the power consumption prediction rule to obtain the predicted power consumption, including:

[0071] The current power consumption feature of length M is obtained, and the current power consumption feature is identified using the power consumption prediction rule to obtain the predicted power consumption corresponding to the future time period.

[0072] In an embodiment of the present invention, based on the predicted power consumption, determining the first-level electric energy consumption warning result corresponding to the target power supply area includes:

[0073] Determine the maximum power generation of all generators in the target power supply area in the future period;

[0074] Determine whether the maximum power generation in the future period is greater than the predicted power consumption corresponding to the future period. If so, determine that the first-level power energy consumption warning result is sufficient power energy; otherwise, determine that the first-level power energy consumption warning result is insufficient power energy.

[0075] In the embodiment of the present invention, when the first-level electric energy consumption warning result is that the electric energy is sufficient, the predicted power consumption is used as a constraint, and the minimum electric energy consumption is determined by using an intelligent optimization algorithm, including:

[0076] When the first-level electric energy consumption warning result is that the electric energy is sufficient, construct an objective function for determining the minimum electric energy consumption amount and construct a target constraint based on the predicted power consumption; wherein the objective function is used to characterize the cost function of all generator sets, and the target constraint is used to characterize the operating constraint of the generator set;

[0077] Optionally, the objective function for minimum power energy consumption is:

[0078]

[0079] Where f represents the objective function, represents the first power generation cost coefficient of the dth generator set, represents the second generation cost coefficient of the dth generator set, represents the third generation cost coefficient of the dth generator set, D represents the total number of generator sets, represents the planned power generation of the dth generator set in the future time period, Indicates the on / off status of the dth generator set in the future time period, and is set to 1 when the generator set is on and set to 0 when it is off; It indicates the startup cost of the dth generator set in the future time period. It can be a hot start cost or a cold start cost, depending on whether the generator set is started in the previous period.

[0080] Construct target constraints based on predicted electricity consumption, including:

[0081] Constraint 1: The sum of the planned power generation of all generators (i.e., power generation in the future) is greater than or equal to the predicted power consumption;

[0082] Constraint 2: The planned power generation of each generator set is between its corresponding upper and lower limits;

[0083] It is worth noting that in addition to these constraints, other constraints can also be set to ensure the normal operation of the power system, such as active power balance constraints. After the above objective function and objective constraints are combined, it is actually the cost function required to generate the predicted power consumption. Finding the minimum cost function can obtain the minimum power energy consumption in the future period.

[0084] Based on the target constraint, the power generation of all power generating units in the future is initialized to obtain a power generation code, and multiple power generation codes are repeatedly obtained;

[0085] Since the power generation in the future of the objective function is in hours, the power generation of each generator set in the next 24 hours can be encoded and the optimal combination can be found to determine the minimum power energy consumption.

[0086] According to the objective function, determine the objective function value corresponding to each power generation code;

[0087] Based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, and the update is repeated until the total number of updates reaches the preset maximum number of updates, and the power generation code with the minimum objective function value is obtained;

[0088] Decode the power generation code with the minimum objective function value to determine the final power generation of each generator set at a future time;

[0089] The pre-stored amount of electric energy consumption required for a unit of power generation is obtained, and the minimum amount of electric energy consumption is determined based on the amount of electric energy consumption required for a unit of power generation and the final power generation of each generator set in the future.

[0090] In the embodiment of the present invention, based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, including:

[0091] A1. Determine a first update ratio and a second update ratio, and divide all power generation codes into a first code group and a second code group according to the first update ratio and the second update ratio;

[0092] A2. Using the best neighborhood exploration strategy to update the first power generation code in the first coding population, to obtain an updated first power generation code;

[0093] A2.1. For any first power generation code, determine its first neighboring code and second neighboring code as follows:

[0094]

[0095]

[0096]

[0097] in, Indicates k The first neighborhood code corresponding to the first power generation code d Dimension parameters, Indicates k The first power generation code corresponds to the second neighborhood code d Dimension parameters, k =1,2,…,K, K Indicates the total number of the first power generation code, d =1,2,…,D, where D represents the total number of generator sets in the target power supply area. Indicates k The first power generation code d Dimension parameters, express The corresponding update speed, represents the search distance in the tth update process, represents the search distance during the t-1th update process, represents the attenuation factor;

[0098] A2.2. Based on the first neighborhood code and the second neighborhood code, the best neighborhood search for the kth first power generation code is:

[0099]

[0100]

[0101] in, Indicates the first power generation code after an update The d-th dimension parameter of represents the inertia weight of the kth first power generation code, represents the first learning factor, represents the second learning factor, represents the first random number between (0,1), represents the second random number between (0,1), represents the historical best position corresponding to the kth first power generation code, Indicates the optimum power generation amount code among all the first power generation amount codes.

[0102] By updating the first power generation code once, the preferred direction of the neighborhood of each first power generation code can be quickly explored, thereby realizing a fast search.

[0103] Optionally, an adaptively changing inertia weight and learning factor are provided:

[0104]

[0105]

[0106] in, Indicates the preset maximum value of inertia weight. Indicates the preset maximum number of updates. Indicates k The objective function value corresponding to the first power generation code, represents the maximum objective function value corresponding to the first power generation code, represents pi, and cos represents the cosine function.

[0107] By setting the above-mentioned adaptively changing inertia weights and learning factors, the algorithm can perform irregular variable speed search, which is more conducive to exploring more unknown areas and speeding up the search speed. In the later stage of the algorithm, it can help to improve the search accuracy, so as to find the parameter combination that minimizes the objective function.

[0108] A3, using a multi-information interaction strategy to update the second power generation code in the second coding population to obtain an updated second power generation code;

[0109] A3.1. Determine the search code corresponding to the nth second power generation code in the second coding population: ;in, represents the d-th dimension parameter in the n-th second power generation encoding;

[0110] A3.2. Based on the search code, determine the code update amount as follows: ;in, represents the search control factor, and , represents the constant coefficient term, v represents the adaptive coefficient, and , represents the upper limit of the adaptive coefficient, represents the lower limit of the adaptive coefficient, e represents a natural constant, Represents a random number in the range [0, 2𝜋] that follows a uniform distribution.

[0111] A3.3. According to the code update amount, the nth second power generation amount code is updated as follows: + ;in, Indicates the second power generation code after an update The d-th dimension parameter in ;

[0112] By updating the second power generation code once as described above, all second power generation codes can be searched in a vortex shape around the optimal power generation code, which not only helps to improve the accuracy of the solution, but also helps to explore more unknown areas, thereby achieving rapid search.

[0113] A4. Use a global greedy search strategy to update the first power generation code after one update and the second power generation code after one update, obtain the first power generation code and the second power generation code after two updates, and complete an update process for each power generation code.

[0114] A4.1. Determine the global search value of the first power generation code after one update and the second power generation code after one update:

[0115]

[0116] in, represents the dth dimension of the i-th power generation code to be globally searched, the power generation code to be globally searched represents the first power generation code after one update or the second power generation code after one update, express The search value of represents the global search control coefficient, , represents the standard Gaussian variation, represents the standard Cauchy variation;

[0117] A4.2. Determine whether the loss function value of the global search value is reduced. If so, accept the update; otherwise, reject the update.

[0118] By performing the global greedy search on the first power generation code after one update and the second power generation code after one update, the algorithm can always have the ability to jump out of the local optimum and ensure the search speed of the algorithm.

[0119] Optionally, the inferior solution acceptance probability can be set. When the loss function value of the global search value does not decrease, a roulette wheel is used to determine whether to accept the update based on the inferior solution acceptance probability. After each update, the parameters of each dimension can be processed for out-of-bounds to ensure that the constraints are met.

[0120] The embodiment of the present invention adopts an intelligent optimization algorithm to update each power generation code, and will not fall into the local optimum. It can quickly find the power generation code with the lowest power generation cost in the high-dimensional solution space, thereby obtaining the minimum power energy consumption, so as to achieve more accurate power energy consumption warning.

[0121] In the embodiment of the present invention, determining the number of days during which electric energy can be consumed based on the minimum electric energy consumption amount includes:

[0122] Determine the current remaining power energy consumption corresponding to the target power supply area and obtain the replenishable power energy consumption corresponding to the future period; wherein the replenishable power energy consumption is obtained from the procurement system of the target power supply area;

[0123] The current remaining power energy consumption corresponding to the target power supply area is added to the replenishable power energy consumption, and then the minimum power energy consumption is subtracted to obtain the comprehensive remaining power energy consumption;

[0124] The number of days that electricity energy can be consumed is obtained by dividing the comprehensive remaining electricity energy consumption by the minimum electricity energy consumption.

[0125] In an embodiment of the present invention, based on the number of days that the electric energy can be consumed, determining the second-level electric energy consumption warning result corresponding to the target power supply area includes:

[0126] Based on the number of days in which the electric energy can be consumed, determine whether the number of days in which the electric energy can be consumed is greater than a preset time threshold. If so, determine that the second-level electric energy consumption warning result is sufficient electric energy; otherwise, determine that the second-level electric energy consumption warning result is insufficient electric energy.

[0127] In the embodiment of the present invention, after the electric energy consumption warning information is generated, the method further includes: transmitting the electric energy consumption warning information to a device or information system designated by a staff member.

[0128] like Figure 2 As shown, the embodiment of the present invention provides an electric energy consumption warning system, including: a prediction module 201, a first warning module 202, a consumable quantity determination module 203, a second warning module 204 and a warning information generation module 205;

[0129] The prediction module 201 is used to learn historical power consumption characteristics using a deep learning algorithm to obtain power consumption prediction rules, and predict power consumption in future time periods based on the power consumption prediction rules to obtain predicted power consumption;

[0130] The first warning module 202 is used to determine the first-level power energy consumption warning result corresponding to the target power supply area based on the predicted power consumption; wherein the first-level power energy consumption warning result includes sufficient power energy or insufficient power energy;

[0131] The consumables determination module 203 is used to determine the minimum power energy consumption by using the predicted power consumption as a constraint and an intelligent optimization algorithm when the first-level power energy consumption warning result is that the power energy is sufficient;

[0132] The second warning module 204 is used to determine the number of days that electric energy can be consumed based on the minimum electric energy consumption, and determine the second-level electric energy consumption warning result corresponding to the target power supply area based on the number of days that electric energy can be consumed; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy;

[0133] The warning information generation module 205 is used to generate power energy consumption warning information when the first-level power energy consumption warning result or the second-level power energy consumption warning result is insufficient power energy, and transmit the power energy consumption warning information to a designated device or information system.

[0134] An electric energy consumption early warning system provided in an embodiment of the present invention can execute the above method embodiment, and its principles and beneficial effects are similar, which will not be described in detail here.

[0135] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and interpreted scripting language JavaScript, etc.

[0136] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0137] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0138] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0139] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0140] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for early warning of power energy consumption, characterized in that: include: After learning the historical electricity consumption characteristics using a deep learning algorithm, an electricity consumption prediction rule is obtained, and based on the electricity consumption prediction rule, the electricity consumption in future time periods is predicted to obtain the predicted electricity consumption; Based on the predicted power consumption, determining the first-level power energy consumption warning result corresponding to the target power supply area; wherein the first-level power energy consumption warning result includes sufficient power energy or insufficient power energy; When the first-level power energy consumption warning result is that the power energy is sufficient, the predicted power consumption is used as a constraint and an intelligent optimization algorithm is used to determine the minimum power energy consumption; Based on the minimum amount of electric energy consumption, the number of days that electric energy can be consumed is determined, and based on the number of days that electric energy can be consumed, the second-level electric energy consumption warning result corresponding to the target power supply area is determined; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy; When the first-level electric energy consumption warning result or the second-level electric energy consumption warning result is insufficient electric energy, electric energy consumption warning information is generated; When the first-level power energy consumption warning result is that the power energy is sufficient, the predicted power consumption is used as a constraint, and an intelligent optimization algorithm is used to determine the minimum power energy consumption, including: When the first-level electric energy consumption warning result is that the electric energy is sufficient, construct an objective function for determining the minimum electric energy consumption amount and construct a target constraint based on the predicted power consumption; wherein the objective function is used to characterize the cost function of all generator sets, and the target constraint is used to characterize the operating constraint of the generator set; Based on the target constraint, the power generation of all power generating units in the future is initialized to obtain a power generation code, and multiple power generation codes are repeatedly obtained; According to the objective function, determine the objective function value corresponding to each power generation code; Based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, and the update is repeated until the total number of updates reaches the preset maximum number of updates, and the power generation code with the minimum objective function value is obtained; Decode the power generation code with the minimum objective function value to determine the final power generation of each generator set at a future time; Obtain the pre-stored amount of electric energy consumption required for a unit of power generation, and determine the minimum amount of electric energy consumption based on the amount of electric energy consumption required for a unit of power generation and the final power generation of each generator set in the future; Based on the objective function value corresponding to each power generation code, an intelligent optimization algorithm is used to update each power generation code, including: A1. Determine a first update ratio and a second update ratio, and divide all power generation codes into a first code group and a second code group according to the first update ratio and the second update ratio; A2. Using the best neighborhood exploration strategy to update the first power generation code in the first coding population, to obtain an updated first power generation code; A2.

1. For any first power generation code, determine its first neighboring code and second neighboring code as follows: in, Indicates k The first neighborhood code corresponding to the first power generation code d Dimension parameters, Indicates k The first power generation code corresponds to the second neighborhood code d Dimension parameters, k =1,2,…,K, K Indicates the total number of the first power generation code, d =1,2,…,D, where D represents the total number of generator sets in the target power supply area. Indicates k The first power generation code d Dimension parameters, express The corresponding update speed, represents the search distance in the tth update process, represents the search distance during the t-1th update process, represents the attenuation factor; A2.

2. Based on the first neighborhood code and the second neighborhood code, the best neighborhood search for the kth first power generation code is: in, Indicates the first power generation code after an update The d-th dimension parameter of represents the inertia weight of the kth first power generation code, represents the first learning factor, represents the second learning factor, represents the first random number between (0,1), represents the second random number between (0,1), represents the historical best position corresponding to the kth first power generation code, represents the optimal power generation code among all first power generation codes; The inertia weight and learning factor of adaptive change are: in, Indicates the preset maximum value of inertia weight. Indicates the preset maximum number of updates. Indicates k The objective function value corresponding to the first power generation code, represents the maximum objective function value corresponding to the first power generation code, represents pi, and cos represents the cosine function; A3, using a multi-information interaction strategy to update the second power generation code in the second coding population to obtain an updated second power generation code; A3.

1. Determine the search code corresponding to the nth second power generation code in the second coding population: ;in, represents the d-th dimension parameter in the n-th second power generation encoding; A3.

2. Based on the search code, determine the code update amount as follows: ;in, represents the search control factor, and , represents the constant coefficient term, v represents the adaptive coefficient, and , represents the upper limit of the adaptive coefficient, represents the lower limit of the adaptive coefficient, e represents a natural constant, represents a random number in the range [0, 2𝜋] that follows a uniform distribution; A3.

3. According to the code update amount, the nth second power generation amount code is updated as follows: + ;in, Indicates the second power generation code after an update The d-th dimension parameter in ; A4. Use a global greedy search strategy to update the first power generation code after the first update and the second power generation code after the first update, obtain the first power generation code and the second power generation code after the second update, and complete the update process of each power generation code once; A4.

1. Determine the global search value of the first power generation code after one update and the second power generation code after one update: in, represents the dth dimension of the i-th power generation code to be globally searched, the power generation code to be globally searched represents the first power generation code after one update or the second power generation code after one update, express The search value of represents the global search control coefficient, , represents the standard Cauchy variation, represents the standard Gaussian variation; A4.

2. Determine whether the loss function value of the global search value is reduced. If so, accept the update; otherwise, reject the update.

2. The electric energy consumption early warning method according to claim 1 is characterized in that: After using the deep learning algorithm to learn the historical electricity consumption characteristics, the electricity consumption prediction rules are obtained, including: Obtain the historical power consumption characteristics corresponding to the target power supply area and the power consumption corresponding to the historical power consumption characteristics; Determine the historical electricity consumption characteristics corresponding to M time points as sample data, and use the electricity consumption corresponding to the M+1th time point as label data; A deep learning algorithm is used to learn the relationship between sample data and label data to obtain electricity consumption prediction rules.

3. The electric energy consumption early warning method according to claim 2 is characterized in that: The power consumption in the future period is predicted based on the power consumption prediction rule to obtain the predicted power consumption, including: The current power consumption feature of length M is obtained, and the current power consumption feature is identified using the power consumption prediction rule to obtain the predicted power consumption corresponding to the future time period.

4. The electric energy consumption early warning method according to claim 1 is characterized in that: Based on the predicted power consumption, the first-level power energy consumption warning result corresponding to the target power supply area is determined, including: Determine the maximum power generation of all generators in the target power supply area in the future period; Determine whether the maximum power generation in the future period is greater than the predicted power consumption corresponding to the future period. If so, determine that the first-level power energy consumption warning result is sufficient power energy; otherwise, determine that the first-level power energy consumption warning result is insufficient power energy.

5. The electric energy consumption early warning method according to claim 1 is characterized in that: Based on the minimum amount of power energy consumption, determine the number of days that power energy can be consumed, including: Determine the current remaining power energy consumption corresponding to the target power supply area and obtain the replenishable power energy consumption corresponding to the future period; wherein the replenishable power energy consumption is obtained from the procurement system of the target power supply area; The current remaining power energy consumption corresponding to the target power supply area is added to the replenishable power energy consumption, and then the minimum power energy consumption is subtracted to obtain the comprehensive remaining power energy consumption; The number of days that electricity energy can be consumed is obtained by dividing the comprehensive surplus electricity energy consumption by the minimum electricity energy consumption.

6. The electric energy consumption early warning method according to claim 1 is characterized in that: Based on the number of days that the electric energy can be consumed, the second-level electric energy consumption warning result corresponding to the target power supply area is determined, including: Based on the number of days in which the electric energy can be consumed, determine whether the number of days in which the electric energy can be consumed is greater than a preset time threshold. If so, determine that the second-level electric energy consumption warning result is sufficient electric energy; otherwise, determine that the second-level electric energy consumption warning result is insufficient electric energy.

7. The electric energy consumption early warning method according to claim 1 is characterized in that: After the electric energy consumption warning information is generated, the method further includes: transmitting the electric energy consumption warning information to a device or information system designated by a staff member.

8. An electric energy consumption early warning system, the electric energy consumption early warning system is used to execute the electric energy consumption early warning method according to any one of claims 1 to 7, characterized in that: include: A prediction module, a first warning module, a consumables quantity determination module, a second warning module, and a warning information generation module; The prediction module is used to learn historical power consumption characteristics using a deep learning algorithm to obtain power consumption prediction rules, and predict power consumption in future time periods based on the power consumption prediction rules to obtain predicted power consumption; The first warning module is used to determine the first-level electric energy consumption warning result corresponding to the target power supply area based on the predicted power consumption; wherein the first-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy; The consumables quantity determination module is used to determine the minimum power energy consumables quantity by using the predicted power consumption as a constraint and an intelligent optimization algorithm when the first-level power energy consumption warning result is that the power energy is sufficient; The second warning module is used to determine the number of days that electric energy can be consumed based on the minimum electric energy consumption, and determine the second-level electric energy consumption warning result corresponding to the target power supply area based on the number of days that electric energy can be consumed; wherein the second-level electric energy consumption warning result includes sufficient electric energy or insufficient electric energy; The warning information generation module is used to generate power energy consumption warning information when the first-level power energy consumption warning result or the second-level power energy consumption warning result is insufficient power energy, and transmit the power energy consumption warning information to a designated device or information system.

Images