CN115169994A

CN115169994A - Multi-clean-energy complementary control decision processing method and system

Info

Publication number: CN115169994A
Application number: CN202211076214.8A
Authority: CN
Inventors: 成华; 刘庆洲; 贾雪峰; 王慧敏
Original assignee: Beijing Yuqian Energy Technology Co ltd
Current assignee: Beijing Yuqian Energy Technology Co ltd
Priority date: 2022-09-05
Filing date: 2022-09-05
Publication date: 2022-10-11
Anticipated expiration: 2042-09-05
Also published as: CN115169994B

Abstract

The invention discloses a multi-clean-energy complementary control decision processing method and system. And obtaining current energy information. Prediction information is obtained. Demand information is obtained. And generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity. And obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information. And issuing an instruction according to the condition that the demand is met, and distributing energy. The distribution model includes a predictive neural network, an energy distribution structure, and an energy acquisition distribution structure. The invention realizes intelligent unified management and control by uniformly coordinating local light energy, local wind energy, local biogas and other heat energy sources. The current energy is regulated and controlled by adopting a prediction method and linking with the possibility of future energy, and meanwhile, the electric quantity is jointly intelligently regulated and controlled according to different characteristics and regulation and control prices of the energy.

Description

Multi-clean-energy complementary control decision processing method and system

Technical Field

The invention relates to the technical field of computers, in particular to a method and a system for processing a complementary control decision of multiple clean energy sources.

Background

With the gradual promotion of energy-saving and emission-reducing policies, the energy structure and energy substitution of China are also greatly changed. The original fossil energy is gradually replaced by clean energy such as solar energy, wind energy, air energy, biological energy and the like, but the problems of instability and power grid consumption which are specific to green energy are accompanied.

At present, on the basis of stable energy supply, the main method for solving the problem of instability of green energy is multi-energy complementation, new energy such as solar energy, wind energy, air energy, biological energy and other forms of energy are systematically and reasonably designed, and meanwhile, the existing energy such as fuel gas and the like are used for bottom-preserving supply and peak-shaving regulation and control in an auxiliary mode. The multi-energy complementation is widely applied in the field of building energy conservation, particularly the field of heat supply of communities and parks, but the effect is good or bad, the difference is large, and the method is extremely dependent on the harmony, the rationality and the sharing of the design of each system. And the combination of multiple energy forms has great influence on system load and pipe network flow. After new energy is added, the power generation can be a combined mode of thermal power generation and new energy (solar energy, wind energy, tidal energy and the like), and the electric quantity of the power generation is accumulated and can change along with natural energy such as the sun, wind, thunder, tide and the like to form an extremely unstable phenomenon. The control logics of various energy forms are disordered, and the control logics are mainly independent of each other, so that the control logics cannot be effectively and uniformly allocated and monitored.

Disclosure of Invention

The present invention provides a method and a system for processing a multi-clean energy complementary control decision, so as to solve the above problems in the prior art.

In a first aspect, an embodiment of the present invention provides a method for processing a multi-clean-energy complementary control decision, including:

obtaining current energy information; the current energy information comprises current stored electricity quantity, current solar energy, current wind energy, current biological energy and current fuel;

obtaining prediction information; the prediction information comprises prediction photovoltaic power generation electric quantity and prediction wind power generation electric quantity;

acquiring demand information; the demand information represents a demand electrical load, a demand thermal load, and a demand cold load;

generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity; the unstable electric quantity is the sum of the electric quantity obtained by photovoltaic power generation of the current solar energy and the electric quantity obtained by wind power generation of the current wind energy;

obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information; the demand satisfaction condition comprises a demand satisfaction value and distributed electric quantity;

according to the condition that the demand is met, issuing an instruction to distribute energy;

the distribution model comprises a prediction neural network, an energy distribution structure and an energy acquisition and distribution structure:

the input of the prediction neural network is the prediction information; the input of the energy distribution structure is the output, the unstable electric quantity and the demand information of the prediction neural network; the input of the energy acquisition and distribution structure is demand information, unstable electric quantity and current energy information.

Optionally, obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information, and the demand information, includes:

judging the unstable electric quantity and the demand information to obtain a demand satisfaction value; the requirement satisfaction value is 1, which represents that the unstable electric quantity satisfies the requirement information; the requirement satisfaction value is 0, which indicates that the unstable electric quantity does not meet the requirement information;

if the demand satisfaction value is 1, obtaining the residual electric quantity; the residual electric quantity is the electric quantity remaining after the requirement information is met;

distributing the unstable electric quantity through an energy distribution structure based on the residual electric quantity and the prediction information to obtain distributed electric quantity; the distributed electric quantity comprises a transmission electric quantity, a storage electric quantity and a abandon electric quantity;

if the demand satisfaction value is 0, acquiring the demand residual capacity; the required residual electric quantity is the difference of the required electric quantity minus the unstable electric quantity;

obtaining the used energy information through an energy acquisition and distribution structure based on the demand information, the demand residual capacity and the current energy information; the information on the used energy includes bio-energy, fuel, stored electricity and external delivery electricity.

Optionally, the allocating the unstable electric quantity through an energy allocation structure based on the remaining electric quantity and the prediction information to obtain an allocated electric quantity includes:

inputting the prediction information into a prediction neural network to obtain predicted electric quantity; the predicted electric quantity is the sum of the wind speed predicted electric quantity and the illumination predicted electric quantity; the illumination prediction electric quantity represents an electric quantity by photovoltaic power generation in predicted future N days; the wind speed prediction electric quantity represents the electric quantity generated by wind energy in predicted N days in the future;

if the predicted electric quantity is larger than the electric quantity threshold value, setting a storage value to be 0; if the predicted electric quantity is smaller than the electric quantity threshold value, setting the storage value to be 1 to obtain a predicted electric quantity difference value; the predicted electric quantity difference value is the difference of the electric quantity threshold value minus the predicted electric quantity;

obtaining distribution price information; the distribution price information comprises a storage electric quantity price and a delivery electric quantity price; the price of the transmitted electric quantity is in direct proportion to the power supply framework area; the price of the stored electricity is in direct proportion to the stored electricity;

if the storage value is 0, obtaining the optimal distribution electric quantity based on the distribution price information and the residual electric quantity; the optimal distribution electric quantity comprises distribution transmission electric quantity and distribution abandon electric quantity;

if the storage value is 1, obtaining the optimal distribution electric quantity based on the distribution price information, the predicted electric quantity difference value and the residual electric quantity; the optimal distribution of the electric quantity comprises distribution of storage electric quantity, distribution of delivery electric quantity and distribution of abandonment electric quantity.

Optionally, the training method of the predictive neural network includes:

obtaining a training set; the training set comprises a plurality of training data groups and a plurality of marking data; the training data set comprises a training wind speed data set and a training illumination data set; the training illumination data set comprises a first illumination monitoring data set, a second illumination monitoring data set and a third illumination monitoring data set; the first illumination monitoring data group is training illumination monitoring data of the previous 5 days; the second illumination monitoring data set is training illumination monitoring data 5 days after the first illumination monitoring data set; the third illumination monitoring data group is training illumination monitoring data 5 days after the second illumination monitoring data group; the first illumination monitoring data set comprises a first total illumination intensity, a second total illumination intensity, a third total illumination intensity, a fourth total illumination intensity and a fifth total illumination intensity; the second illumination monitoring data set comprises a sixth total illumination intensity, a seventh total illumination intensity, an eighth total illumination intensity, a ninth total illumination intensity and a tenth total illumination intensity; the third illumination monitoring data set includes an eleventh total illumination intensity, a twelfth total illumination intensity, a thirteenth total illumination intensity, a fourteenth total illumination intensity, and a fifteenth total illumination intensity; the marking data comprise marking illumination electric quantity and marking wind speed electric quantity; the marked illumination electric quantity represents the total power generation amount of photovoltaic power generation for 15 days; the marked wind speed electric quantity represents the total power generation amount of 15 days of wind energy power generation;

inputting the training illumination data set into an illumination prediction neural network to obtain illumination prediction electric quantity;

inputting the training wind speed data set into a wind speed prediction neural network to obtain wind speed prediction electric quantity;

obtaining an illumination loss value through an illumination loss function according to the illumination prediction electric quantity and the marked illumination electric quantity;

the wind speed prediction electric quantity and the marked wind speed electric quantity are used for obtaining a wind speed loss value through a wind speed loss function;

obtaining a total loss value; the total loss value is the sum of the illumination loss value and the wind speed loss value;

obtaining the current training iteration number of a prediction neural network and the preset maximum iteration number of the training of the prediction neural network;

and stopping training when the total loss value is less than or equal to a threshold value or the training iteration number reaches the maximum iteration number, so as to obtain the trained predictive neural network.

Optionally, the inputting the training illumination data set into an illumination prediction neural network to obtain the illumination prediction electric quantity includes:

inputting the third illumination monitoring data group into an LSTM structure according to a time sequence from far to near to obtain third output information; the third output information includes a fifteenth power generation amount, a fourteenth power generation amount, a thirteenth power generation amount, a twelfth power generation amount, and an eleventh power generation amount; the LSTM structures are formed by sequentially connecting 5 LSTMs;

inputting the second illumination monitoring data group and the eleventh generated energy into an LSTM structure according to a time sequence from far to near to obtain second output information; the second output information includes tenth power generation amount, ninth power generation amount, eighth power generation amount, seventh power generation amount, and sixth power generation amount; the LSTM structures are formed by sequentially connecting 5 LSTMs;

inputting the first illumination monitoring data group and the sixth power generation amount into an LSTM structure according to a time sequence from far to near to obtain first output information; the first output information comprises a fifth power generation amount, a fourth power generation amount, a third power generation amount, a second power generation amount and a first power generation amount; the LSTM structures are formed by sequentially connecting 5 LSTMs;

and adding the first power generation amount, the second power generation amount, the third power generation amount, the fourth power generation amount, the fifth power generation amount, the sixth power generation amount, the seventh power generation amount, the eighth power generation amount, the ninth power generation amount, the tenth power generation amount, the eleventh power generation amount, the twelfth power generation amount, the thirteenth power generation amount, the fourteenth power generation amount and the fifteenth power generation amount to obtain illumination prediction power.

Optionally, the inputting the first illumination monitoring data set and the sixth power generation amount into the LSTM structure according to a time sequence from far to near to obtain first output information includes:

inputting the fifth total illumination intensity and the sixth power generation amount into a fifth LSTM to obtain fifth LSTM information; the fifth LSTM information includes a fifth power generation amount and a fifth LSTM output value;

inputting the fourth total illumination intensity and the fifth LSTM output value into a fourth LSTM to obtain fourth LSTM information; the fourth LSTM information includes a fourth power generation amount and a fourth LSTM output value;

inputting the third total illumination intensity and the fourth LSTM output value into a third LSTM to obtain third LSTM information; the third LSTM information includes a third power generation amount and a third LSTM output value;

inputting the second total illumination intensity and the third LSTM output value into a second LSTM to obtain second LSTM information; the second LSTM information includes a second power generation amount and a second LSTM output value;

inputting the first total illumination intensity and the second LSTM output value into a first LSTM to obtain first LSTM information; the first LSTM information is a first power generation amount.

Optionally, if the storage value is 0, obtaining an optimal distribution electric quantity based on the distribution price information, the predicted electric quantity difference value, and the remaining electric quantity, includes:

acquiring the required transmission electric quantity; the required transmission electric quantity sent by each current region of the required transmission electric quantity is transmitted;

the optimal distribution electric quantity is obtained by the following formula calculation method:

Q=X*MoneyX+Y1* MoneyY1+Y2* MoneyY2+ Y3* MoneyY3+Z*Money；

X+Y1+Y2+Y3+Z=K；

X<=Xmax； Y1<=Y1max；Y2<=Y2max；Y3<=Y3max；

wherein Q is the lowest price; x is stored electricity quantity; moneyX is the price of the stored electricity quantity and is a constant; y1 is the first area for transmitting electric quantity; moneyY1 is the price of the electric quantity transported by the first area and is a constant; y2 is the transmission electric quantity of the second area; moneyY2 is the price of the electric quantity conveyed by the second area and is a constant; y3 is the third area transmission electric quantity; moneyY3 is the price of the electric quantity transmitted by the third area and is a constant; z is the electricity abandoning quantity; the Money is the price for generating electricity again by using the biological energy after electricity is abandoned and is a constant; k is the residual capacity; xmax is the predicted electric quantity difference; y1max is the required electric quantity for conveying the first area; y2max is the required electric quantity for conveying the second area; and Y3max is the required electric quantity transmitted by the third area.

Optionally, obtaining the used energy information through an energy acquisition and distribution structure based on the demand information, the demand remaining capacity, and the current energy information includes:

generating power, heating and refrigerating through combined cooling heating and power equipment based on bioenergy to obtain bioenergy information; the bioenergy information comprises bioenergy electrical load, bioenergy thermal load and bioenergy cold load;

if the biological energy information meets the requirement information, the redundant biological energy information is stored with electric quantity;

if the biological energy does not meet the requirement information, based on the biological energy information, fuel and stored electricity are adopted, and fuel information is obtained through combined cooling heating and power generation equipment; the fuel information indicates the amount of fuel injected.

Optionally, if the bioenergy is for unsatisfied demand information, based on the bioenergy information, adopt fuel and storage electric quantity, through the combined cooling heating and power equipment, obtain fuel information, include:

performing electric heating and electric refrigeration on the stored electric quantity to obtain stored electric quantity conversion information; the stored electric quantity conversion information comprises a stored electric quantity conversion value and a stored electric quantity conversion value; the conversion value of the stored electricity quantity is 1, which indicates that the stored electricity discharge can meet the demand information; the conversion value of the stored electricity quantity is 0, which indicates that the stored electricity discharge can not meet the demand information;

when the stored electricity conversion value is 1, the stored electricity conversion value is the electricity used for meeting the demand information; when the stored electricity conversion value is 0, converting the stored electricity conversion value into stored electricity;

if the conversion value of the stored electric quantity is 0, using fuel to obtain the consumed fuel quantity through combined cooling heating and power equipment; the consumed fuel amount is the amount of fuel that meets the demand.

In a second aspect, an embodiment of the present invention provides a system for processing a multi-clean-energy complementary control decision, including:

an acquisition module: obtaining current energy information; the current energy information comprises current stored electricity quantity, current solar energy, current wind energy, current biological energy and current fuel; obtaining prediction information; the prediction information comprises prediction photovoltaic power generation electric quantity and prediction wind power generation electric quantity; acquiring demand information; the demand information indicates a demand electrical load, a demand thermal load, and a demand cooling load;

the light energy and wind energy power generation calculation module: generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity; the unstable electric quantity is the sum of the electric quantity obtained by photovoltaic power generation of the current solar energy and the electric quantity obtained by wind power generation of the current wind energy;

a distribution module: obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information; the demand satisfaction condition comprises a demand satisfaction value and distributed electric quantity; according to the condition that the demand is met, issuing an instruction to distribute energy;

Compared with the prior art, the embodiment of the invention achieves the following beneficial effects:

the embodiment of the invention also provides a method and a system for processing the complementary control decision of multiple clean energy sources, wherein the method comprises the following steps: and obtaining current energy information. The current energy information includes a current stored electricity amount, a current solar energy, a current wind energy, a current bio-energy, and a current fuel. Prediction information is obtained. The prediction information comprises prediction photovoltaic power generation electric quantity and prediction wind power generation electric quantity. Demand information is obtained. The demand information indicates a demand electric load, a demand heat load, and a demand cold load. Generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity; the unstable electric quantity is the sum of the electric quantity obtained by photovoltaic power generation of the current solar energy and the electric quantity obtained by wind power generation of the current wind energy. And obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information. The demand satisfaction condition comprises a demand satisfaction value and distributed electric quantity. And issuing an instruction according to the condition that the demand is met, and distributing energy. The distribution model comprises a prediction neural network, an energy distribution structure and an energy acquisition and distribution structure: the input of the predictive neural network is the predictive information. The inputs of the energy distribution structure are the output of the predictive neural network, the unstable electric quantity and the demand information. The input of the energy acquisition and distribution structure is demand information, unstable electric quantity and current energy information.

The invention relates to an integrated and modularized intelligent decision making technology and a system device, which can be matched with the characteristics of photovoltaic, wind power, fuel gas and other multi-energy systems. The invention realizes intelligent unified management and control by uniformly coordinating local light energy, local wind energy, local biogas and other heat energy sources. And (3) adopting a prediction method, relating to the possibility of future energy, designing an accurate neural network according to a time sequence to obtain a predicted value, and regulating and controlling the current energy. Meanwhile, a function is constructed according to the energy regulation and control price, and simultaneously different characteristics of energy are combined to jointly intelligently regulate and control the electric quantity.

Drawings

Fig. 1 is a flowchart of a multi-clean-energy complementary control decision processing method according to an embodiment of the present invention;

fig. 2 is a schematic diagram of a power generation, heat generation and refrigeration structure of a multi-clean-energy complementary control decision processing system according to an embodiment of the present invention;

fig. 3 is a schematic diagram of a predictive neural network structure in a multi-clean-energy complementary control decision processing system according to an embodiment of the present invention;

fig. 4 is a schematic block structure diagram of an electronic device according to an embodiment of the present invention.

The labels in the figure are: a bus 500; a receiver 501; a processor 502; a transmitter 503; a memory 504; a bus interface 505.

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings.

Example 1

As shown in fig. 1, an embodiment of the present invention provides a method for processing a multi-clean energy complementary control decision, where the method includes:

s101: and obtaining current energy information. The current energy information includes a current stored power amount, a current solar energy, a current wind energy, a current biological energy, and a current fuel.

Wherein the current energy information represents a current energy situation.

S102: prediction information is obtained. The prediction information comprises prediction photovoltaic power generation electric quantity and prediction wind power generation electric quantity.

S103: and acquiring the demand information. The demand information indicates a demand electrical load, a demand thermal load, and a demand cooling load.

S104: and generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity. The unstable electric quantity is the sum of the electric quantity obtained by photovoltaic power generation of the current solar energy and the electric quantity obtained by wind power generation of the current wind energy.

Wherein the energy generating, heat generating and cooling structure is shown in figure 2.

S105: and obtaining a demand meeting condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information. The demand satisfaction condition comprises a demand satisfaction value and distributed electric quantity.

Wherein the requirement satisfaction value of 1 indicates that the requirement information is satisfied. The requirement satisfaction value of 0 indicates that the requirement information is not satisfied.

S106: and issuing an instruction according to the condition that the demand is met, and distributing energy.

the schematic diagram of the predictive neural network structure is shown in fig. 3.

By the method, the provincial electric power and electric quantity demand, the section conveying capacity and the channel conveying capacity are boundaries, and the demands in the peak period and the low peak period are intelligently distributed, so that the generated electric quantity and heat energy are distributed, and the optimal distribution method meeting the energy demand can be obtained.

and judging the unstable electric quantity and the demand information to obtain a demand satisfaction value. The requirement satisfaction value of 1 represents unstable electric quantity satisfaction requirement information. The requirement satisfaction value of 0 indicates that the unstable electric quantity does not satisfy the requirement information.

And if the demand satisfaction value is 1, obtaining the residual electric quantity. The residual electric quantity is the electric quantity remaining after the requirement information is met.

The electric quantity obtained by photovoltaic power generation and wind power generation is subtracted, and the sum of the electric quantity meeting the heat load demand, the electric quantity meeting the cold load demand and the electric quantity meeting the electric load demand is added to obtain the residual electric quantity.

Distributing the unstable electric quantity through an energy distribution structure based on the residual electric quantity and the prediction information to obtain distributed electric quantity; the distributing the electric quantity comprises transmitting the electric quantity, storing the electric quantity and abandoning the electric quantity.

And if the demand satisfaction value is 0, obtaining the demand residual electric quantity. The required residual electric quantity is the difference of the required electric quantity minus the unstable electric quantity.

The unstable electric quantity meets the requirement heat load through electric heating, the electric quantity after meeting the requirement cold load through electric refrigeration is smaller than the requirement electric load, and the requirement information is not met by the unstable electric quantity.

Obtaining the used energy information through an energy obtaining and distributing structure based on the demand information, the demand residual capacity and the current energy information; the information on the used energy includes bio-energy, fuel, stored electricity and external delivery electricity.

With the above method, since wind energy and solar energy are particularly dependent on weather, the amount of electricity to generate electricity often jumps between meeting and not meeting the demand information. Because the satisfied operation is different from the unsatisfied operation, the residual electric quantity is distributed and calculated through the judgment instruction, or the obtained distributed electric quantity is calculated to meet the demand information.

and inputting the prediction information into a prediction neural network to obtain the predicted electric quantity. The predicted electric quantity is the sum of the wind speed predicted electric quantity and the illumination predicted electric quantity. The illumination prediction electric quantity represents an electric quantity by photovoltaic power generation in predicted future N days. The wind speed prediction electric quantity represents an electric quantity generated by wind energy on predicted future N days.

And if the predicted electric quantity is larger than the electric quantity threshold value, setting the storage value to be 0. If the predicted electric quantity is smaller than the electric quantity threshold value, setting the storage value to be 1 to obtain a predicted electric quantity difference value; the predicted electric quantity difference value is the difference of the electric quantity threshold value minus the predicted electric quantity.

The electric quantity threshold value is the total electric quantity which is provided from the current time to the previous 15 days and can meet the normal work of the current area.

Allocation price information is obtained. The distribution price information includes a stored electricity price and a delivered electricity price. The price of the stored electricity is in direct proportion to the stored electricity. The price of the transmitted electric quantity is in direct proportion to the area of the power supply framework.

Wherein a power supply skeleton region is constructed. The power supply framework region is a power supply circuit and is divided into different regions according to the power supply distance. The power supply line is a line which takes the current area as the center and supplies power to other areas according to the actual life. The price of the transmitted electric quantity is in direct proportion to the power supply framework area.

And if the storage value is 0, obtaining the optimal distribution electric quantity based on the distribution price information, the predicted electric quantity difference value and the residual electric quantity. The optimal distribution of the electric quantity comprises distribution of the electric quantity for delivery and distribution of the electric quantity for abandonment.

And if the storage value is 1, obtaining the optimal distribution electric quantity based on the distribution price information, the predicted electric quantity difference value and the residual electric quantity. The optimal distribution of the electric quantity comprises distribution of storage electric quantity, distribution of delivery electric quantity and distribution of abandonment electric quantity.

By the method, if the predicted electric quantity is larger than the electric quantity threshold value, the photovoltaic power generation and the wind power generation in the next few days can meet the requirement of the ordinary electric quantity, so that the electric quantity does not need to be stored by wasting resources, and the residual electric quantity is selectively transmitted or discarded. If the electric quantity is excessive and the power supply is selected to be transmitted outwards under judgment, a power supply route is obtained, an elliptical partition line is drawn according to the power supply distance, the elliptical partition line is divided into different areas according to different partition lines, and power is supplied to different power supply areas. Since different prices for power transmission are divided according to the output distance, the areas are divided according to the transmission distance of the power supply route which is different from the area, and the power supply route is allocated to different areas. If the predicted electric quantity is smaller than the electric quantity threshold value, the common electric quantity requirement is not met in the next few days, storage is needed, and the upper limit of the storage is the predicted electric quantity difference value. The remaining power is selectively stored, transported or discarded.

Optionally, the training method of the predictive neural network includes:

a training set is obtained. The training set includes a plurality of training data sets and a plurality of label data. The training data set comprises a training wind speed data set and a training illumination data set. The training illumination data set comprises a first illumination monitoring data set, a second illumination monitoring data set and a third illumination monitoring data set. The first lighting monitoring data set is the lighting training monitoring data of the previous 5 days. The second illumination monitoring data set is training illumination monitoring data 5 days after the first illumination monitoring data set; the third illumination monitoring data group is training illumination monitoring data 5 days after the second illumination monitoring data group; the first illumination monitoring data set comprises a first total illumination intensity, a second total illumination intensity, a third total illumination intensity, a fourth total illumination intensity and a fifth total illumination intensity; the second illumination monitoring data set includes a sixth total illumination intensity, a seventh total illumination intensity, an eighth total illumination intensity, a ninth total illumination intensity, and a tenth total illumination intensity; the third lighting monitoring data set includes an eleventh total lighting intensity, a twelfth total lighting intensity, a thirteenth total lighting intensity, a fourteenth total lighting intensity, and a fifteenth total lighting intensity. The annotation data indicates the total amount of electricity generated for 15 days. And the marking data comprise marking illumination electric quantity and marking wind speed electric quantity. The marked illumination electric quantity represents the total power generation amount of photovoltaic power generation for 15 days. The marked wind speed electric quantity represents the total power generation amount of 15 days of wind energy power generation.

Wherein the values in the training lighting data set represent a total lighting intensity for 24 hours; the values in the training wind speed data set represent a total wind speed of 24 hours. The training wind speed data set comprises a first wind speed monitoring data set, a second wind speed monitoring data set and a third wind speed monitoring data set; the first wind speed monitoring data group is training wind speed monitoring data of the previous 5 days; the second wind speed monitoring data group is training wind speed monitoring data 5 days after the first wind speed monitoring data group; the third wind speed monitoring data group is training wind speed monitoring data 5 days after the second wind speed monitoring data group; the first wind speed monitoring data group comprises a first total wind speed intensity, a second total wind speed intensity, a third total wind speed intensity, a fourth total wind speed intensity and a fifth total wind speed intensity; the second wind speed monitoring data group comprises a sixth total wind speed intensity, a seventh total wind speed intensity, an eighth total wind speed intensity, a ninth total wind speed intensity and a tenth total wind speed intensity; the third wind speed monitoring data set includes an eleventh total wind speed intensity, a twelfth total wind speed intensity, a thirteenth total wind speed intensity, a fourteenth total wind speed intensity and a fifteenth total wind speed intensity.

And inputting the training illumination data set into an illumination prediction neural network to obtain illumination prediction electric quantity.

And inputting the training wind speed data set into a wind speed prediction neural network to obtain wind speed prediction electric quantity.

The wind speed prediction neural network structure is the same as the illumination prediction neural network structure, but the parameters obtained by training are different due to two different inputs, so that two networks are adopted for prediction.

And obtaining the illumination loss value through an illumination loss function according to the illumination prediction electric quantity and the marked illumination electric quantity.

Wherein the illumination loss function is a binary cross entropy loss function.

And (4) obtaining a wind speed loss value through a wind speed loss function by using the wind speed prediction electric quantity and the marked wind speed electric quantity.

Wherein the wind speed loss function is a binary cross entropy loss function.

Obtaining a total loss value; the total loss value is the sum of the illumination loss value and the wind speed loss value.

And obtaining the current training iteration number of the prediction neural network and the preset maximum iteration number of the training of the prediction neural network.

Wherein, the number of times is 12000 in the embodiment.

And stopping training when the total loss value is less than or equal to the threshold value or the training iteration number reaches the maximum iteration number, so as to obtain the trained predictive neural network.

By the method, the electric quantity used at the time before the current time is adopted for judgment because the demand of the electric quantity has great relation with seasons. Since weather forecasts can only be roughly predicted, the longer the time the less influential, the LSTM is used.

inputting the third illumination monitoring data group into an LSTM structure according to a time sequence from far to near to obtain third output information; the third output information includes a fifteenth power generation amount, a fourteenth power generation amount, a thirteenth power generation amount, a twelfth power generation amount, and an eleventh power generation amount; the LSTM structures are 5 LSTMs which are connected in sequence.

Inputting the second illumination monitoring data group and the eleventh generated energy into an LSTM structure according to a time sequence from far to near to obtain second output information; the second output information includes a tenth power generation amount, a ninth power generation amount, an eighth power generation amount, a seventh power generation amount, and a sixth power generation amount; the LSTM structures are 5 LSTMs which are connected in sequence.

Inputting the first illumination monitoring data group and the sixth generating capacity into an LSTM structure according to a time sequence from far to near to obtain first output information; the first output information comprises a fifth power generation amount, a fourth power generation amount, a third power generation amount, a second power generation amount and a first power generation amount; the LSTM structures are 5 LSTMs which are connected in sequence.

With the above method, since there is a time relationship between predicted weather and thus between energy sources generated by weather, the time relationship is retained by predicting using LSTM. However, since the input data is too long for fifteen days, if the data for fifteen days is directly input into the LSTM at one time, the data information input at the beginning will be reduced continuously during the continuous operation of the network. But the relationship between days was consistent. Dividing the days into three data with 5 days as length according to time sequence, establishing LTSM structure circulation composed of 5 LSTMs for detection, and inputting 5 data each time, not only inputting predicted energy information, but also inputting the electric quantity obtained by the last day of the previous input data as a parameter. Because the input is the electric quantity information of which the last predicted energy output is direct, the last input information can be combined, the reduction of information characteristics caused by direct one-time input of 15 data is avoided, and the electric quantity can be obtained more accurately.

inputting the fifth total illumination intensity and the sixth power generation amount into a fifth LSTM to obtain fifth LSTM information; the fifth LSTM information includes a fifth power generation amount and a fifth LSTM output value.

Inputting the fourth total illumination intensity and the fifth LSTM output value into a fourth LSTM to obtain fourth LSTM information; the fourth LSTM information includes a fourth power generation amount and a fourth LSTM output value.

Inputting the third total illumination intensity and the fourth LSTM output value into a third LSTM to obtain third LSTM information; the third LSTM information includes a third power generation amount and a third LSTM output value.

Inputting the second total illumination intensity and the third LSTM output value into a second LSTM to obtain second LSTM information; the second LSTM information includes a second power generation amount and a second LSTM output value.

By the method, the network is designed, the electric quantity generated by the current illumination intensity is output every time the LSTM is input, and the electric quantity is input in a reverse order, so that the relationship of the illumination intensity is tighter when the time is shorter, and more accurate detection total is obtained.

acquiring the required transmission electric quantity; and the required transmission electric quantity sent by each current region is transmitted.

Q=X*MoneyX+Y1* MoneyY1+Y2* MoneyY2+ Y3* MoneyY3+Z*Money；

X+Y1+Y2+Y3+Z=K；

X<=Xmax； Y1<=Y1max；Y2<=Y2max；Y3<=Y3max；

wherein Q is the lowest price. X is stored electricity quantity; moneyX is the price of the stored electricity quantity and is a constant. Y1 is the first area for transmitting electric quantity. MoneyY1 is the price of the electric quantity transported in the first area and is a constant. Y2 is the transmission electric quantity of the second area; moneyY2 is the price of the electric quantity transported by the second area and is a constant. Y3 is the third area transmission electric quantity; and MoneyY3 is the price of the electric quantity transmitted by the third area and is a constant. And Z is the electricity abandoning quantity. Money is a constant price for electricity regeneration using biological energy after electricity abandonment; and K is the residual capacity. Xmax is the predicted charge difference. Y1max is required electric quantity transmitted by the first region; and Y2max is the required electric quantity delivered by the second area. And Y3max is the required electric quantity delivered by the third area.

The first area is used for transmitting the required electric quantity, wherein the required electric quantity is required by the first area to request other areas. The second area transmits the required electric quantity, namely the electric quantity required by the second area to request other areas. The third area transmits the required electric quantity to the other areas requested by the third area.

In this embodiment, the price of stored electricity is 1.5 yuan/watt hour. MoneyY1, namely the price of the electric quantity transported by the first area is 0.8 yuan/watt hour. MoneyY2, namely the price of the electricity transmission amount in the second area is 1.2 yuan/watt hour. The price of the electricity delivered by the third area is 1.8 yuan/watt hour.

By the above method, the minimum value is obtained by the derivative function. And according to the lowest price, the second low-price method distributes the electric quantity in sequence for calculation. And carrying out operations of conveying, storing or abandoning the electricity and the like on the near electricity quantity. Drawing an elliptical separation line according to the power supply distance through a power supply route, dividing the elliptical separation line into three areas according to different separation lines, and supplying power to different power supply areas. Since different prices caused in the electric transportation are divided according to the output distance, the three areas have different prices. And combining the mode of distributing the electric quantity with the price required by each mode, and finding the optimal distributed electric quantity according to the lowest price.

Optionally, obtaining the used energy information through an energy acquisition and distribution structure based on the demand information, the demand remaining power and the current energy information includes:

generating power, heating and refrigerating by using bioenergy and through combined cooling heating and power equipment to obtain bioenergy information; the bioenergy information comprises a bioenergy electrical load, a bioenergy thermal load and a bioenergy cold load.

And if the biological energy information meets the requirement information, the redundant biological energy information is stored with electric quantity.

The biological energy satisfying requirement information is generated by heating and electricity through combined cooling heating and power generation equipment, and the biological energy electricity load after cooling and electricity cooling through the combined cooling heating and power equipment is larger than the electricity load of the required residual electricity.

By the above method, since the stored value is discharged by storing first. The bioenergy power generation is determined by the amount of fuel generated by the living beings, and the bioenergy power generation amount is stable. Because the energy of the biological energy power generation can be regenerated, whether the biological energy power generation meets the requirement is considered, and if the biological energy power generation does not meet the requirement, the method of reusing the stored electric quantity for discharging is used. Since the stored electricity is discharged and then the stored electricity is not satisfied, when the bio-energy is used, the life of the storage device is consumed in the process of rapid discharging and charging because the industrial design may generate the surplus electricity, such as storing the surplus electricity.

Optionally, if the bioenergy is unsatisfied demand information, based on the bioenergy information, adopt fuel and storage electric quantity, through the combined cooling heating and power equipment, obtain fuel information, include:

performing electric heating and electric refrigeration on the stored electric quantity to obtain stored electric quantity conversion information; the stored electric quantity conversion information comprises a stored electric quantity conversion value and a stored electric quantity conversion value; the conversion value of the stored electricity quantity is 1, which indicates that the stored electricity discharge can meet the requirement information; the conversion value of the stored electricity quantity is 0, which indicates that the stored electricity discharge can not meet the requirement information.

And when the stored electric quantity conversion value is 1, the stored electric quantity conversion value is the electric quantity used for meeting the demand information. And when the stored electricity conversion value is 0, converting the stored electricity into stored electricity.

The storage electric quantity discharge can meet the requirement that the sum of the heat load generated by the storage electric quantity discharge heating and the biological energy heat load is equal to the requirement heat load, the sum of the cold load generated by the storage electric quantity discharge refrigeration and the biological energy cold load is equal to the requirement cold load, and the sum of the electric load generated by the storage electric quantity discharge and the biological energy electric load is equal to the requirement electric load. When the stored electricity conversion value is 1, the stored electricity conversion value is the electricity discharged by satisfying the above condition. The stored electricity conversion amount with the stored electricity conversion value of 0 is all the stored electricity and also all the discharged electricity.

And if the sum of the conversion values of the stored electric quantity is 0, using fuel to obtain the consumed fuel quantity through combined cooling heating and power equipment. The consumed fuel amount is the amount of fuel that meets the demand.

The fuel quantity is the fuel quantity of the sum of the biological energy heat load and the biological energy heat load equal to the required heat load, the fuel cooling load and the storage electric quantity discharge refrigeration generated cooling load generated by the combined cooling, heating and power generation equipment, the biological energy cooling load and the biological energy cooling load equal to the required cooling load, the fuel electric load and the storage electric quantity discharge electric load generated by the combined cooling, heating and power generation equipment and the biological energy electric load equal to the required electric load.

With the above method, the fuel is the most polluting energy source of nature, and is therefore reused at last.

By the method, green energy is a great trend, and people have already entered the green energy era, and the green energy such as wind power, photovoltaic and the like is developed rapidly in recent years. However, the green energy has the characteristics of volatility, randomness, inverse peak regulation and the like, and the safe and stable operation of a power grid can be influenced by large-scale grid connection. The problems of grid connection difficulty, low absorption rate and the like are caused, and even phenomena of wind abandonment, light abandonment and the like are caused. The green energy AI comes from delivery, and the storage, the transport and the abandonment of surplus electric quantity are judged by using prediction, and the electric quantity is intelligently regulated and controlled to be supplemented when the electric quantity is insufficient, so that the research on how to give the urban electric meters to the wind, rain and thunder and lightning of the nature is relieved.

Example 2

Based on the above complementary control decision processing method for multiple clean energy sources, the embodiment of the invention also provides a complementary control decision processing system for multiple clean energy sources, wherein the system comprises an acquisition module, a light energy and wind energy power generation calculation module and a distribution module.

The acquisition module is used for acquiring current energy information. The current energy information includes a current stored electricity amount, a current solar energy, a current wind energy, a current bio-energy, and a current fuel. Obtaining prediction information; the prediction information comprises prediction photovoltaic power generation electric quantity and prediction wind power generation electric quantity; demand information is obtained. The demand information indicates a demand electric load, a demand heat load, and a demand cold load.

And the light energy and wind energy power generation calculation module is used for generating power by using the current solar energy and the current wind energy to obtain unstable electric quantity. The unstable electric quantity is the sum of the electric quantity obtained by photovoltaic power generation of the current solar energy and the electric quantity obtained by wind power generation of the current wind energy.

The distribution module is used for obtaining a demand meeting condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information. The demand satisfaction condition comprises a demand satisfaction value and an allocation electric quantity. And issuing an instruction according to the condition that the demand is met, and distributing energy.

By adopting the scheme, the electric quantity generated subsequently is predicted, the specific prediction is realized through the model of the scheme, the time length of input data is long, the loss amount of the hidden information of the data predicted by the traditional LSTM network is large, the data is leased for the purpose, the circulating LSTM network is designed, the problem of the hidden information loss caused by the overlong data length is solved, and the prediction accuracy is improved.

The specific manner in which the respective modules perform operations has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

An embodiment of the present invention further provides an electronic device, as shown in fig. 4, which includes a memory 504, a processor 502 and a computer program stored in the memory 504 and executable on the processor 502, wherein the processor 502 implements the steps of any one of the methods for processing complementary control decision for clean energy described above when executing the program.

Where in fig. 4 a bus architecture (represented by bus 500) is shown, bus 500 may include any number of interconnected buses and bridges, and bus 500 links together various circuits including one or more processors, represented by processor 502, and memory, represented by memory 504. The bus 500 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 505 provides an interface between the bus 500 and the receiver 501 and transmitter 503. The receiver 501 and the transmitter 503 may be the same element, i.e. a transceiver, providing a means for communicating with various other apparatus over a transmission medium. The processor 502 is responsible for managing the bus 500 and general processing, and the memory 504 may be used for storing data used by the processor 502 in performing operations.

Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of any one of the above-mentioned methods for processing complementary control decisions of clean energy and the related data.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, the present invention is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: rather, the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore, may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components in an apparatus according to an embodiment of the invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.

Claims

1. A multi-clean energy complementary control decision processing method is characterized by comprising the following steps:

obtaining current energy information; the current energy information comprises current stored electricity, current solar energy, current wind energy, current biological energy and current fuel;

acquiring demand information; the demand information indicates a demand electrical load, a demand thermal load, and a demand cooling load;

2. The method as claimed in claim 1, wherein obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the forecast information and the demand information comprises:

if the demand satisfaction value is 0, obtaining the demand residual electric quantity; the required residual electric quantity is the difference of the required electric quantity minus the unstable electric quantity;

3. The method as claimed in claim 2, wherein the allocating unstable electric power to obtain allocated electric power by an energy allocation structure based on the remaining electric power and the prediction information comprises:

4. The method as claimed in claim 3, wherein the training method of the predictive neural network comprises:

obtaining a training set; the training set comprises a plurality of training data groups and a plurality of marking data; the training data set comprises a training wind speed data set and a training illumination data set; the training illumination data set comprises a first illumination monitoring data set, a second illumination monitoring data set and a third illumination monitoring data set; the first illumination monitoring data group is training illumination monitoring data of the previous 5 days; the second illumination monitoring data set is training illumination monitoring data 5 days after the first illumination monitoring data set; the third illumination monitoring data group is training illumination monitoring data 5 days after the second illumination monitoring data group; the first illumination monitoring data set comprises a first total illumination intensity, a second total illumination intensity, a third total illumination intensity, a fourth total illumination intensity and a fifth total illumination intensity; the second illumination monitoring data set includes a sixth total illumination intensity, a seventh total illumination intensity, an eighth total illumination intensity, a ninth total illumination intensity, and a tenth total illumination intensity; the third illumination monitoring data set includes an eleventh total illumination intensity, a twelfth total illumination intensity, a thirteenth total illumination intensity, a fourteenth total illumination intensity, and a fifteenth total illumination intensity; the marking data comprise marking illumination electric quantity and marking wind speed electric quantity; the marked illumination electric quantity represents the total power generation amount of photovoltaic power generation for 15 days; the marked wind speed electric quantity represents the total power generation amount of 15 days of wind energy power generation;

acquiring the current training iteration number of a prediction neural network and the preset maximum iteration number of the training of the prediction neural network;

5. The method as claimed in claim 4, wherein the inputting the training illumination data set into an illumination prediction neural network to obtain an illumination prediction electric quantity comprises:

inputting the third illumination monitoring data set into an LSTM structure according to a time sequence from far to near to obtain third output information; the third output information includes a fifteenth power generation amount, a fourteenth power generation amount, a thirteenth power generation amount, a twelfth power generation amount, and an eleventh power generation amount; the LSTM structures are formed by sequentially connecting 5 LSTMs;

inputting the second illumination monitoring data group and the eleventh generated energy into an LSTM structure according to a time sequence from far to near to obtain second output information; the second output information includes a tenth power generation amount, a ninth power generation amount, an eighth power generation amount, a seventh power generation amount, and a sixth power generation amount; the LSTM structures are formed by sequentially connecting 5 LSTMs;

6. The method as claimed in claim 5, wherein the inputting the first illumination monitoring data set and the sixth power generation amount into the LSTM structure in time sequence from far to near to obtain the first output information comprises:

7. The method as claimed in claim 3, wherein if the stored value is 0, obtaining an optimal distribution power based on the distribution price information, the predicted power difference value and the remaining power comprises:

Q=X*MoneyX+Y1* MoneyY1+Y2* MoneyY2+ Y3* MoneyY3+Z*Money；

X+Y1+Y2+Y3+Z=K；

X<=Xmax； Y1<=Y1max；Y2<=Y2max；Y3<=Y3max；

wherein Q is the lowest price; x is the stored electricity quantity; moneyX is the price of the stored electricity quantity and is a constant; y1 is the first area for transmitting electric quantity; moneyY1 is the price of the electric quantity transported by the first area and is a constant; y2 is the second area transmission electric quantity; moneyY2 is the price of the electric quantity transported by the second area and is a constant; y3 is the third area transmission electric quantity; moneyY3 is the price of the electric quantity transmitted by the third area and is a constant; z is electricity abandonment quantity; money is a constant price for electricity regeneration using biological energy after electricity abandonment; k is the residual capacity; xmax is the predicted electric quantity difference; y1max is required electric quantity transmitted by the first region; y2max is the required electric quantity for conveying the second area; and Y3max is the required electric quantity transmitted by the third area.

8. The method as claimed in claim 2, wherein the obtaining of the used energy information through an energy acquisition and distribution structure based on the demand information, the demand remaining capacity and the current energy information comprises:

generating power, heating and refrigerating through combined cooling heating and power equipment based on bioenergy to obtain bioenergy information; the bioenergy information comprises a bioenergy electrical load, a bioenergy thermal load and a bioenergy cold load;

if the bioenergy is the information which does not meet the requirements, fuel information is obtained by adopting fuel and stored electricity through combined cooling heating and power equipment based on the bioenergy information; the fuel information indicates the amount of fuel injected.

9. The method as claimed in claim 8, wherein if the bio-energy is not enough to meet the requirement, the obtaining the fuel information through the combined cooling, heating and power generation equipment by using the fuel and the stored electricity based on the bio-energy information comprises:

performing electric heating and electric refrigeration on the stored electric quantity to obtain stored electric quantity conversion information; the stored electric quantity conversion information comprises a stored electric quantity conversion value and a stored electric quantity conversion quantity; the conversion value of the stored electricity quantity is 1, which indicates that the stored electricity discharge can meet the demand information; the conversion value of the stored electricity quantity is 0, which indicates that the stored electricity discharge can not meet the demand information;

when the stored electric quantity conversion value is 1, the stored electric quantity conversion value is the electric quantity used for meeting the demand information; when the stored electricity conversion value is 0, converting the stored electricity conversion value into stored electricity;

10. A multi-clean-energy complementary control decision processing system is characterized by comprising:

a distribution module: obtaining a demand satisfaction condition through a distribution model based on the unstable electric quantity, the current energy information, the prediction information and the demand information; the demand satisfaction condition comprises a demand satisfaction value and distributed electric quantity; issuing an instruction according to the condition that the demand is met, and distributing energy;