An optimization scheduling method and terminal based on probability box and conditional risk value

Abstract

The present invention discloses an optimization scheduling method and terminal based on probability box and conditional risk value, which determine the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period; determine the corresponding cumulative probability density curve based on each prediction error, and determine the interval set that meets the preset confidence from the cumulative probability density curve based on each prediction error according to the probability box theory; calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence based on the conditional risk value theory; construct an optimization scheduling model based on the final uncertainty factor fluctuation interval, and use an intelligent optimization algorithm to solve the optimization scheduling model to obtain an optimization scheduling plan, thereby ensuring the integrity of uncertain information, calculating uncertainty situations outside the confidence interval, avoiding the conservatism of traditional interval optimization methods while ensuring the sufficiency of safety constraints, and achieving an effective balance between the economy and safety of the distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network, and in particular to an optimization scheduling method and a terminal based on probability box and conditional risk value. Background Art

[0002] With the "dual carbon" goal being proposed and the power system reform being continuously promoted, the access of large-scale renewable energy generators to the power grid has become an important development trend of the power system. Among them, the access of renewable energy generators to the distribution network in a distributed and small-capacity form is an important mode, but the randomness and volatility of distributed renewable energy (RDG) may lead to many problems such as voltage over-limit, line congestion and increased operation and maintenance costs.

[0003] The deterministic power flow calculation method is difficult to adapt to the economic dispatch of power systems under complex uncertain conditions, and it is not suitable to manage and control high-proportion renewable energy power systems based on it. In view of natural factors such as wind speed and light intensity that affect RDG output, establishing an interval optimization model based on the interval power flow calculation method can effectively characterize its uncertainty, which is a relatively common uncertainty optimization method.

[0004] The essence of interval optimization method is to use interval arithmetic to solve the best combination of decision variables so that the target can be optimized in interval form. In the process of interval boundary extraction, the traditional interval method uses the worst scenario as the interval boundary. To ensure safety, the width of the generated uncertainty fluctuation interval is too large, resulting in poor economic efficiency of the optimization result. The interval truncation method achieves the purpose of improving the economy of the optimization result by compressing the interval width by pre-setting the confidence level. However, since it cannot take into account the uncertain factors outside the confidence interval, it has the disadvantage of being difficult to ensure its safety. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an optimization scheduling method and terminal based on probability box and conditional risk value, which can achieve an effective balance between the economy and safety of the distribution network.

[0006] In order to solve the above technical problems, a technical solution adopted by the present invention is:

[0007] An optimization scheduling method based on probability box and conditional risk value includes the following steps:

[0008] Determine the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period;

[0009] Determine a corresponding cumulative probability density function curve based on each of the prediction errors, and determine a set of intervals that meet a preset confidence level from the cumulative probability density curve based on each of the prediction errors according to the probability box theory;

[0010] Calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence level based on the conditional risk value theory;

[0011] An optimization scheduling model is constructed based on the fluctuation range of the final uncertain factors, and an intelligent optimization algorithm is used to solve the optimization scheduling model to obtain an optimization scheduling solution.

[0012] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0013] An optimization scheduling terminal based on probability box and conditional risk value includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0014] Determine the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period;

[0015] Determine a corresponding cumulative probability density curve based on each of the prediction errors, and determine a set of intervals satisfying a preset confidence level from the cumulative probability density curve based on each of the prediction errors according to the probability box theory;

[0016] Calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence level based on the conditional risk value theory;

[0017] An optimization scheduling model is constructed based on the fluctuation range of the final uncertain factors, and an intelligent optimization algorithm is used to solve the optimization scheduling model to obtain an optimization scheduling solution.

[0018] The beneficial effects of the present invention are as follows: a corresponding cumulative probability density curve is determined based on each prediction error, and a set of intervals satisfying a preset confidence level is determined from the cumulative probability density curve based on each prediction error according to the probability box theory, a final uncertainty factor fluctuation interval corresponding to the set of intervals satisfying the preset confidence level is calculated based on the conditional risk value theory, and an optimization scheduling model is constructed based on the final uncertainty factor fluctuation interval, thereby using the probability box theory to describe the random uncertainty of distributed renewable energy output and load demand and the cognitive uncertainty of model parameters to ensure the integrity of uncertain information, and using the conditional risk value theory based on the uncertainty set to calculate uncertainty situations outside the confidence interval, thereby avoiding the conservatism of traditional interval optimization methods while ensuring the sufficiency of safety constraints, so that the final optimization scheduling scheme achieves an effective balance between the economy and safety of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of the steps of an optimization scheduling method based on probability box and conditional risk value according to an embodiment of the present invention;

[0020] Figure 2 It is a structural schematic diagram of an optimization scheduling terminal based on probability box and conditional risk value according to an embodiment of the present invention;

[0021] Figure 3 This is a flowchart of an optimization scheduling method based on probability box and conditional risk value in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in conjunction with the implementation modes and the accompanying drawings.

[0023] Please refer to Figure 1 The embodiment of the present invention provides an optimization scheduling method based on probability box and conditional risk value, comprising the steps of:

[0024] Determine the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period;

[0025] Determine a corresponding cumulative probability density curve based on each of the prediction errors, and determine a set of intervals satisfying a preset confidence level from the cumulative probability density curve based on each of the prediction errors according to the probability box theory;

[0026] Calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence level based on the conditional risk value theory;

[0027] An optimization scheduling model is constructed based on the fluctuation range of the final uncertain factors, and an intelligent optimization algorithm is used to solve the optimization scheduling model to obtain an optimization scheduling solution.

[0028] From the above description, it can be seen that the beneficial effects of the present invention are: determining the corresponding cumulative probability density curve based on each prediction error, and determining a set of intervals that meet the preset confidence from the cumulative probability density curve based on each prediction error according to the probability box theory, calculating the final uncertainty factor fluctuation interval corresponding to the set of intervals that meet the preset confidence based on the conditional risk value theory, and constructing an optimization scheduling model based on the final uncertainty factor fluctuation interval, thereby using the probability box theory to describe the random uncertainty of distributed renewable energy output and load demand and the cognitive uncertainty of model parameters to ensure the integrity of uncertain information, and using the conditional risk value theory based on the uncertainty set to calculate uncertainty situations outside the confidence interval, while avoiding the conservatism of traditional interval optimization methods while ensuring the sufficiency of safety constraints, so that the final optimized scheduling scheme achieves an effective balance between the economy and safety of the distribution network.

[0029] Furthermore, the step of determining the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period includes:

[0030] Obtaining a first historical data actual value of a distributed renewable energy output and a corresponding first historical data predicted value and a second historical data actual value of a load demand and a corresponding second historical data predicted value within a preset period;

[0031] Determine a first prediction error of the output of the distributed renewable energy source according to the actual value of the first historical data and the corresponding predicted value of the first historical data;

[0032] A second prediction error of the load demand is determined according to the actual value of the second historical data and the corresponding predicted value of the second historical data.

[0033] From the above description, it can be seen that the prediction error is determined according to the actual value of historical data and the predicted value of historical data, which facilitates the subsequent determination of the probability distribution parameter change range based on the prediction error.

[0034] Further, determining the corresponding cumulative probability density curve based on each prediction error includes:

[0035] Determining a probability distribution type corresponding to each of the prediction errors;

[0036] Determine a corresponding probability distribution parameter variation interval according to the probability distribution type;

[0037] The corresponding cumulative probability density curve is determined according to the probability distribution parameter variation interval.

[0038] From the above description, it can be seen that the largest graph formed by the envelope of all cumulative probability density curves represents the uncertainty set of the prediction error. This set takes into account both the random uncertainty of the prediction error and the cognitive uncertainty in the model parameter fitting process, thus ensuring the integrity of the uncertainty factors.

[0039] Further, the probability distribution type includes a normal distribution type;

[0040] Determining the corresponding probability distribution parameter variation interval according to the probability distribution type includes:

[0041] According to the normal distribution type, the corresponding probability distribution parameter variation interval is determined as:

[0042]

[0043] Where x represents the prediction error type, represents the mean of the normal distribution of the prediction error, represents the standard deviation of the normal distribution of the forecast error, represents the lower limit value corresponding to the mean, represents the upper limit value corresponding to the mean, represents the lower limit value corresponding to the standard deviation, represents the upper limit value corresponding to the standard deviation;

[0044] The cumulative probability density curve is:

[0045]

[0046] In the formula, Indicates the upper left boundary of the probability box corresponding to the prediction error, Indicates the upper right boundary of the probability box corresponding to the prediction error, Indicates the lower left boundary of the probability box corresponding to the prediction error, Indicates the lower right boundary of the probability box corresponding to the prediction error, N(a,(b) 2 ) indicates that the curve follows a normal distribution with a as the expectation and b as the standard deviation.

[0047] From the above description, it can be seen that the probability distribution parameters can be combined into multiple cumulative probability density curves, and the probability box is formed by the envelope of each cumulative probability density curve.

[0048] Further, the determining of a set of intervals satisfying a preset confidence level from the cumulative probability density curve based on each prediction error according to the probability box theory includes:

[0049] Determine a prediction error interval that satisfies a preset confidence level based on the prediction error and the preset confidence level according to the probability box theory;

[0050] A set of intervals satisfying a preset confidence level is determined from the cumulative probability density curve according to the prediction error intervals satisfying a preset confidence level.

[0051] Furthermore, the prediction error interval that meets the preset confidence level is:

[0052]

[0053] In the formula, represents the upper boundary of the forecast error interval, represents the lower boundary of the forecast error interval, Indicates the lower right boundary of the probability box corresponding to the prediction error, It represents the upper left boundary of the probability box corresponding to the prediction error, and α represents the preset confidence.

[0054] From the above description, it can be seen that within the probability box, each prediction error corresponds to a cumulative probability. Under the condition of preset confidence, when the difference between the cumulative probabilities corresponding to the upper and lower boundaries of a prediction error interval is not less than the preset confidence, it is the prediction error interval that meets the preset confidence, so as to determine the set of intervals that meet the preset confidence from the cumulative probability density curve based on the prediction error interval that meets the preset confidence.

[0055] Furthermore, the final uncertainty factor fluctuation interval corresponding to the interval set satisfying the preset confidence level is calculated based on the conditional value at risk theory, including:

[0056] Determine the upper and lower bounds of the forecast error under the preset confidence level based on the risk value theory;

[0057] Calculate the probability box-CVaR interval corresponding to the interval set that meets the preset confidence according to the prediction error upper boundary and the prediction error lower boundary;

[0058] The probability box-CVaR interval with the smallest interval width is determined as the final uncertainty factor fluctuation interval.

[0059] Furthermore, the upper boundary of the prediction error under the preset confidence for:

[0060]

[0061] In the formula, represents a certain type of prediction error, e represents the preset threshold, represents the probability that the prediction error does not exceed the preset threshold, α represents the preset confidence, ξ represents the random variable of the prediction error, p(ξ) represents the probability density function corresponding to the random variable of the prediction error, The threshold constraint representing the upper bound of the prediction error;

[0062] The lower boundary of the prediction error under the preset confidence for:

[0063]

[0064] In the formula, represents the probability that the prediction error exceeds the preset threshold, The threshold constraint representing the lower bound of the prediction error;

[0065] The probability box-CVaR interval is:

[0066]

[0067] In the formula, represents the upper boundary of the probability box-CVaR interval, Represents the lower boundary of the probability box-CVaR interval;

[0068] The final uncertainty fluctuation range satisfy:

[0069]

[0070] From the above description, it can be seen that the initial risk value theory obtains the upper and lower boundaries of the interval that meet the confidence level under the condition of a given confidence level. However, this method is similar to the interval truncation method and does not take into account the uncertainty factors outside the confidence interval. Therefore, the subsequent part further uses the conditional risk value theory to consider the uncertainty situations corresponding to the interval that does not meet the confidence level, so as to ensure the safety of the distribution network.

[0071] Furthermore, constructing an optimization scheduling model based on the final uncertainty factor fluctuation range includes:

[0072] Constructing an objective function and constraints according to the fluctuation range of the final uncertain factors, wherein the constraints include power flow constraints of the distribution network, line transmission capacity constraints, node voltage constraints, and controllable distributed power supply operation constraints;

[0073] Obtaining an optimized scheduling model according to the objective function and constraint conditions;

[0074] The objective function minf is:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080] In the formula, represents the electricity purchase cost of the distribution network at time t, represents the gas turbine operating cost at time t, represents the charging and discharging cost of the energy storage system at time t, represents the compensation cost of interruptible load shedding at time t, represents the electricity purchase price of the distribution network at time t, θ represents the θ order of the interval number, It represents the maximum amount of electricity purchased by the distribution network from the upper power grid at time t. N represents the minimum value of the amount of electricity purchased by the distribution network from the upper grid at time t. MT represents the set of nodes with gas turbines in the system, represents the unit operating cost of the gas turbine at node i, represents the power generation of the gas turbine at node i at time t, N ESS Represents the set of nodes with energy storage systems in the system. represents the unit operating cost of the energy storage system at node j, α j,t Represents variable, SOC j,t It represents the state of charge of the energy storage system at node j at time t, SOC j,t+1 represents the state of charge of the energy storage system at node j at time t+1, N IL It indicates the set of nodes with interruptible load in the system. represents the compensation cost for cutting off the interruptible load per unit active load at node k, represents the electricity price of the distribution network at time t, It represents the active power of the interruptible load removal at node k at time t;

[0081] The power flow constraint of the distribution network is:

[0082]

[0083] In the formula, represents the variation range of the active power injected into node i at time t, represents the variation range of the active load injected at node i at time t, represents the node voltage variation range at node i at time t, represents the node voltage variation range at node j at time t, G ij represents the conductance between lines ij, θ ij,t represents the voltage phase angle variation interval between nodes i and j at time t, B ij represents the susceptance between lines ij, represents the variation range of reactive power injected at node i at time t, represents the variation range of reactive load injected at node i at time t, and n represents the number of nodes in the distribution network;

[0084] The line transmission capacity constraint is:

[0085]

[0086] In the formula, It represents the minimum value of active power actually transmitted by line l at time t. It represents the maximum value of active power actually transmitted by line l at time t. Indicates the maximum active power allowed to be transmitted by line l;

[0087] The node voltage constraint is:

[0088]

[0089] In the formula, (U i,min ,U i,max ) represents the voltage fluctuation range allowed at node i;

[0090] The controllable distributed power supply operation constraints are:

[0091]

[0092] In the formula, represents the minimum power generation of the gas turbine at node i, represents the maximum power generation of the gas turbine at node i, represents the maximum downward ramp rate of the gas turbine at node i, △t represents the time interval between two adjacent scheduling times, represents the power generation of the gas turbine at node i at time t-1, represents the maximum upward ramp rate of the gas turbine at node i, SOC j,min represents the minimum state of charge at node j, SOC j,max represents the maximum state of charge at node j, represents the maximum discharge rate of the energy storage system at node j, represents the maximum charging rate of the energy storage system at node j, SOC j,0 It represents the state of charge of the energy storage system at node j at the beginning of the scheduling cycle, SOC j,T represents the charge state of the energy storage system at node j at the end of the scheduling period, It represents the maximum active power that can be removed by the interruptible load at node k.

[0093] Please refer to Figure 2 Another embodiment of the present invention provides an optimization scheduling terminal based on probability box and conditional risk value, including a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the above-mentioned optimization scheduling method based on probability box and conditional risk value is implemented.

[0094] The above-mentioned optimization scheduling method and terminal based on probability box and conditional risk value of the present invention can be applied to the distribution network that needs to balance economy and safety, and is described below through specific implementation methods:

[0095] Embodiment 1

[0096] Please refer to Figure 1 and Figure 3 , an optimization scheduling method based on probability box and conditional risk value in this embodiment includes the steps of:

[0097] S1. Determine the prediction errors corresponding to the output of distributed renewable energy and load demand within a preset period, such as Figure 3 As shown, specifically including:

[0098] S11, obtaining a first historical data actual value of a distributed renewable energy output and a corresponding first historical data predicted value, and a second historical data actual value of a load demand and a corresponding second historical data predicted value within a preset period;

[0099] Wherein, the distributed renewable energy (Renewable Distributed Generation, RDG) includes wind power (Wind power Generation, WG) units and photovoltaic units, etc.;

[0100] S12. Determine a first prediction error of the output of the distributed renewable energy according to the actual value of the first historical data and the corresponding predicted value of the first historical data;

[0101] S13, determining a second prediction error of the load demand according to the actual value of the second historical data and the corresponding predicted value of the second historical data;

[0102] S2. Determine a corresponding cumulative probability density curve based on each of the prediction errors, and determine a set of intervals that meet a preset confidence level from the cumulative probability density curve based on each of the prediction errors according to the probability box theory, such as Figure 3 As shown, specifically including:

[0103] S21, determining the probability distribution type corresponding to each of the prediction errors, where the significance of determining the probability distribution type is to clarify the number and type of distribution parameters, so as to facilitate describing the probability box boundary with the help of the variation range of the distribution parameters;

[0104] In an optional embodiment, the probability distribution type includes a normal distribution type. In another optional embodiment, the probability distribution type also includes a uniform distribution and a Cauchy distribution, etc.;

[0105] S22, determining a corresponding probability distribution parameter variation interval according to the probability distribution type;

[0106] In an optional implementation, the corresponding probability distribution parameter variation interval is determined according to the normal distribution type as follows:

[0107]

[0108] Where x represents the prediction error type, represents the mean of the normal distribution of the prediction error, represents the standard deviation of the normal distribution of the forecast error, represents the lower limit value corresponding to the mean, represents the upper limit value corresponding to the mean, represents the lower limit value corresponding to the standard deviation, represents the upper limit value corresponding to the standard deviation;

[0109] S23, determining a corresponding cumulative probability density curve according to the probability distribution parameter variation interval;

[0110] In an optional implementation, the cumulative probability density curve is:

[0111]

[0112] In the formula, Indicates the upper left boundary of the probability box corresponding to the prediction error, Indicates the upper right boundary of the probability box corresponding to the prediction error, Indicates the lower left boundary of the probability box corresponding to the prediction error, Indicates the lower right boundary of the probability box corresponding to the prediction error, N(a,(b) 2 ) indicates that the curve follows a normal distribution with a as the expectation and b as the standard deviation;

[0113] The probability distribution parameters can be combined into multiple cumulative probability density curves. The probability box is formed by the envelope of each cumulative probability density curve, and its boundary is the cumulative probability density curve.

[0114] S24, determining a prediction error interval that satisfies a preset confidence level based on the prediction error and the preset confidence level according to the probability box theory;

[0115] The prediction error interval that meets the preset confidence level is:

[0116]

[0117] In the formula, represents the upper boundary of the forecast error interval, represents the lower boundary of the forecast error interval, Indicates the lower right boundary of the probability box corresponding to the prediction error, Indicates the upper left boundary of the probability box corresponding to the prediction error, and α indicates the preset confidence;

[0118] S25, determining a set of intervals satisfying a preset confidence level from the cumulative probability density curve according to the prediction error intervals satisfying a preset confidence level;

[0119] S3. Calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence level based on the conditional risk value theory, such as Figure 3 As shown, specifically including:

[0120] S31. Determine the upper boundary and the lower boundary of the prediction error under the preset confidence level based on the risk value theory;

[0121] Among them, the upper boundary of the prediction error under the preset confidence for:

[0122]

[0123] In the formula, represents a certain type of prediction error, e represents the preset threshold, represents the probability that the prediction error does not exceed the preset threshold, α represents the preset confidence, ξ represents the random variable of the prediction error, p(ξ) represents the probability density function corresponding to the random variable of the prediction error, The threshold constraint representing the upper bound of the prediction error;

[0124] The lower boundary of the prediction error under the preset confidence for:

[0125]

[0126] In the formula, represents the probability that the prediction error exceeds the preset threshold, The threshold constraint representing the lower bound of the prediction error;

[0127] S32, calculating the probability box-CVaR interval corresponding to the interval set that meets the preset confidence according to the prediction error upper boundary and the prediction error lower boundary;

[0128] Among them, the probability box-CVaR interval is:

[0129]

[0130] In the formula, represents the upper boundary of the probability box-CVaR interval, Represents the lower boundary of the probability box-CVaR interval;

[0131] S33, determining the probability box-CVaR interval with the smallest interval width as the final uncertainty factor fluctuation interval;

[0132] Among them, the fluctuation range of the final uncertainty factor satisfy:

[0133]

[0134] S4. Constructing an optimization scheduling model based on the fluctuation range of the final uncertain factors, and solving the optimization scheduling model using an intelligent optimization algorithm to obtain an optimization scheduling solution, specifically including:

[0135] S41, constructing an objective function and constraint conditions according to the final uncertainty factor fluctuation range, wherein the constraint conditions include distribution network power flow constraint, line transmission capacity constraint, node voltage constraint and controllable distributed power supply operation constraint;

[0136] Wherein, the objective function minf is:

[0137]

[0138]

[0139]

[0140]

[0141]

[0142] In the formula, represents the electricity purchase cost of the distribution network at time t, represents the operating cost of the MicroTurbine (MT) at time t, represents the charging and discharging cost of the energy storage system (ESS) at time t, represents the compensation cost of interruptible load (IL) removal at time t, represents the purchase price of electricity of the distribution network at time t, θ represents the θ order of the interval number. Since the interval optimization method is adopted in the optimization process of the present invention, the purchase amount of the distribution network from the upper power grid and the purchase fee of the distribution network are both in interval form. In order to compare the advantages and disadvantages of different objective function intervals, the interval form is converted into the corresponding deterministic numerical form by using the θ order relationship of the interval for easy comparison. That is, θ is an auxiliary parameter introduced to facilitate the comparison of the advantages and disadvantages of different objective function intervals, which is used to convert the interval form into the numerical form. In this example, θ takes the value of 0.5. It represents the maximum amount of electricity purchased by the distribution network from the upper power grid at time t. N represents the minimum value of the amount of electricity purchased by the distribution network from the upper grid at time t. MT represents the set of nodes with gas turbines in the system, represents the unit operating cost of the gas turbine at node i, represents the power generation of the gas turbine at node i at time t, N ESS Represents the set of nodes with energy storage systems in the system. represents the unit operating cost of the energy storage system at node j, α j,t It represents a variable, with a value of 0 or 1. 0 means that the energy storage system at node j is in a discharging state at time t, and 1 means that the energy storage system at node j is in a charging state at time t. SOC j,t It represents the state of charge of the energy storage system at node j at time t, SOC j,t+1 represents the state of charge of the energy storage system at node j at time t+1, N IL It indicates the set of nodes with interruptible load in the system. represents the compensation cost for cutting off the interruptible load per unit active load at node k, represents the electricity price of the distribution network at time t, It represents the active power of the interruptible load removal at node k at time t;

[0143] The power flow constraint of the distribution network is:

[0144]

[0145] In the formula, represents the variation range of the active power injected into node i at time t, represents the variation range of the active load injected at node i at time t, represents the node voltage variation range at node i at time t, represents the node voltage variation range at node j at time t, G ij represents the conductance between lines ij, θij,t represents the voltage phase angle variation interval between nodes i and j at time t, B ij represents the susceptance between lines ij, represents the variation range of reactive power injected at node i at time t, represents the variation range of reactive load injected at node i at time t, and n represents the number of nodes in the distribution network;

[0146] The line transmission capacity constraint is:

[0147]

[0148] In the formula, It represents the minimum value of active power actually transmitted by line l at time t. It represents the maximum value of active power actually transmitted by line l at time t. Indicates the maximum active power allowed to be transmitted by line l; when the power transmission direction is from the upstream node to the downstream node, the active power actually transmitted by line l is positive, otherwise it is negative;

[0149] The node voltage constraint is:

[0150]

[0151] In the formula, (U i,min ,U i,max ) represents the voltage fluctuation range allowed at node i;

[0152] The controllable distributed power supply operation constraints are:

[0153]

[0154] In the formula, represents the minimum power generation of the gas turbine at node i, represents the maximum power generation of the gas turbine at node i, represents the maximum downward ramp rate of the gas turbine at node i, △t represents the time interval between two adjacent scheduling times, represents the power generation of the gas turbine at node i at time t-1, represents the maximum upward ramp rate of the gas turbine at node i, SOC j,min represents the minimum state of charge at node j, SOC j,max represents the maximum state of charge at node j, represents the maximum discharge rate of the energy storage system at node j, represents the maximum charging rate of the energy storage system at node j, SOC j,0 It represents the state of charge of the energy storage system at node j at the beginning of the scheduling cycle, SOCj,T represents the charge state of the energy storage system at node j at the end of the scheduling period, represents the maximum active power that can be removed by the interruptible load at node k;

[0155] When the system line parameters, node load demand, operating status of each distributed power source (the power generation of wind turbines, photovoltaic units, gas turbines, the charging and discharging amount of the energy storage system, and the interruptable load interruption amount) and cost coefficient are given, the distribution network line flow can be calculated accordingly, and then the objective function value can be obtained and whether the constraint conditions are met can be judged. However, the output of new energy power generation such as wind and solar power is uncertain, so the present invention adopts the interval method to obtain the corresponding flow state, and considers its uncertainty in the optimization process. The final interval corresponds to the upper and lower boundaries of the corresponding wind power output and photovoltaic output, that is, the load demand, thereby calculating the change interval of the system line flow, and then obtaining the objective function under the uncertainty condition and judging whether the constraint conditions are met, thereby achieving optimization and verifying the effectiveness of the method of the present invention;

[0156] S42, obtaining an optimized scheduling model according to the objective function and constraint conditions;

[0157] S43, using an intelligent optimization algorithm to solve the optimization scheduling model to obtain an optimization scheduling solution;

[0158] Among them, existing intelligent optimization algorithms can be used to solve the optimization scheduling model, such as particle swarm optimization and genetic algorithm, etc., which are not limited here;

[0159] The present invention describes the prediction errors of RDG output and load demand using interval-form distribution parameters based on probability box theory, and takes the probability box of RDG and load demand prediction errors as its uncertainty set. On the basis of interval truncation method, the conditional risk value method is used to take into account the prediction error outside the confidence interval, and the final interval boundary is obtained by the shortest confidence interval principle. At the same time, the power fluctuation interval obtained by the probability box-CVaR method is used as the uncertainty variable and the prediction value is used as the deterministic variable to construct the corresponding distribution network optimization scheduling model, so as to further realize the safe and economical scheduling of the distribution network.

[0160] Since the CVaR method obtains the interval boundary based on data characteristics and corresponding confidence levels, the method has a low reliance on subjective decision-making and can make adaptive changes for different prediction error situations. Its optimization solution achieves a balance between economy and safety, as shown in Tables 1 and 2. Table 1 shows the prediction error intervals generated by different methods, and Table 2 shows the operation results of the optimization solutions corresponding to different methods in 10,000 simulation scenarios.

[0161] The traditional interval method uses the worst scenario as the interval boundary during the interval boundary extraction process. In order to ensure safety, the width of the uncertainty factor fluctuation interval generated by this method is too large, which may lead to poor economic efficiency of the optimization result; the interval truncation method achieves the purpose of improving the economic efficiency of the optimization result by compressing the interval width by pre-setting the confidence level, but it is difficult to ensure its safety because it cannot take into account the uncertain factors outside the confidence interval; on the basis of the interval truncation method, the method of the present invention uses the conditional risk value theory to describe the uncertainty factors outside the confidence interval, and fully takes into account the uncertainty factors outside the confidence level while compressing the interval width. Therefore, as shown in Table 1, The method of the present invention makes targeted improvements to the existing methods, and the generated forecast error intervals of wind power output, photovoltaic output and load demand are between the interval truncation method and the traditional interval method; as shown in Table 2, the probability of over-limit of system line transmission power of the method of the present invention is slightly higher than that of the traditional interval method, but its economic improvement effect is significant, and the main network operation cost is reduced by 10%. The economy of the interval truncation method is equivalent to that of the present invention, but its over-limit risk is 3 times that of the method of the present invention. It can be seen that the method of the present invention avoids the disadvantage of the traditional interval optimization method being too conservative on the one hand, and improves the problem of insufficient safety constraints in the interval truncation method on the other hand.

[0162] Table 1 Prediction error intervals of different methods

[0163]

[0164]

[0165] Table 2 Comparison of optimization effects of different optimization schemes under simulation scenarios

[0166]

[0167] Embodiment 2

[0168] Please refer to Figure 2 In this embodiment, an optimization scheduling terminal based on probability box and conditional risk value includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the optimization scheduling method based on probability box and conditional risk value in the first embodiment is implemented.

[0169] In summary, the present invention provides an optimization scheduling method and terminal based on probability box and conditional risk value, which determine the prediction errors corresponding to the output and load demand of distributed renewable energy within a preset period; determine the corresponding cumulative probability density curve based on each of the prediction errors, and determine the interval set that meets the preset confidence from the cumulative probability density curve based on each of the prediction errors according to the probability box theory; calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence based on the conditional risk value theory; construct an optimization scheduling model based on the final uncertainty factor fluctuation interval, and use an intelligent optimization algorithm to solve the optimization scheduling model to obtain an optimization scheduling plan, thereby using the probability box theory to describe the random uncertainty of the output and load demand of distributed renewable energy and the cognitive uncertainty of the model parameters, ensuring the integrity of the uncertain information, and using the conditional risk value theory based on the uncertain set to calculate the uncertainty outside the confidence interval, while avoiding the conservatism of the traditional interval optimization method while ensuring the sufficiency of the safety constraints, so that the final optimization scheduling plan achieves an effective balance between the economy and safety of the distribution network.

[0170] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An optimization scheduling method based on probability box and conditional risk value, It is characterized in that Includes steps: Determine the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period; Determine a corresponding cumulative probability density curve based on each of the prediction errors, and determine a set of intervals satisfying a preset confidence level from the cumulative probability density curve based on each of the prediction errors according to the probability box theory; Calculate the final uncertainty factor fluctuation interval corresponding to the interval set that meets the preset confidence level based on the conditional risk value theory; Building an optimization scheduling model based on the fluctuation range of the final uncertain factors, and solving the optimization scheduling model using an intelligent optimization algorithm to obtain an optimization scheduling solution; The step of determining the prediction errors corresponding to the output of distributed renewable energy and the load demand in a preset period includes: Obtaining a first historical data actual value of a distributed renewable energy output and a corresponding first historical data predicted value and a second historical data actual value of a load demand and a corresponding second historical data predicted value within a preset period; Determine a first prediction error of the distributed renewable energy output according to the first historical data actual value and the corresponding first historical data predicted value; Determining a second prediction error of the load demand according to the actual value of the second historical data and the corresponding predicted value of the second historical data; Determining the corresponding cumulative probability density curve based on each prediction error includes: Determining a probability distribution type corresponding to each of the prediction errors; Determine a corresponding probability distribution parameter variation interval according to the probability distribution type; Determine the corresponding cumulative probability density curve according to the probability distribution parameter variation interval; The probability distribution type includes a normal distribution type; Determining the corresponding probability distribution parameter variation interval according to the probability distribution type includes: According to the normal distribution type, the corresponding probability distribution parameter variation interval is determined as: ； Where x represents the prediction error type, represents the mean of the normal distribution of forecast errors, represents the standard deviation of the normal distribution of forecast errors, represents the lower limit value corresponding to the mean, represents the upper limit value corresponding to the mean, represents the lower limit value corresponding to the standard deviation, represents the upper limit value corresponding to the standard deviation; The cumulative probability density curve is: ； In the formula, , Respectively represent the upper left and upper right boundaries of the probability box corresponding to the prediction error, , Respectively represent the lower left and lower right boundaries of the probability box corresponding to the prediction error, N (a,(b) 2 ) indicates that the curve follows a normal distribution with a as the expectation and b as the standard deviation; Determining a set of intervals satisfying a preset confidence level from the cumulative probability density curve based on each prediction error according to the probability box theory includes: Determine a prediction error interval that satisfies a preset confidence level based on the prediction error and the preset confidence level according to the probability box theory; Determining a set of intervals satisfying a preset confidence level from the cumulative probability density curve according to the prediction error intervals satisfying a preset confidence level; The final uncertainty factor fluctuation interval corresponding to the interval set satisfying the preset confidence level calculated based on the conditional value at risk theory includes: Determine the upper and lower bounds of the forecast error under the preset confidence level based on the risk value theory; Calculate the probability box-CVaR interval corresponding to the interval set that meets the preset confidence according to the prediction error upper boundary and the prediction error lower boundary; The probability box-CVaR interval with the smallest interval width is determined as the final uncertainty factor fluctuation interval.

2. The optimization scheduling method based on probability box and conditional risk value according to claim 1, It is characterized in that The prediction error interval that meets the preset confidence level is: ； In the formula, represents the upper boundary of the forecast error interval, represents the lower boundary of the forecast error interval, Indicates the lower right boundary of the probability box corresponding to the prediction error, Indicates the upper left boundary of the probability box corresponding to the prediction error, Indicates the preset confidence level.

3. The optimization scheduling method based on probability box and conditional risk value according to claim 1, It is characterized in that The upper boundary of the prediction error under the preset confidence for: ； In the formula, represents a certain type of prediction error, e represents the preset threshold, represents the probability that the prediction error does not exceed the preset threshold, represents the preset confidence level, A random variable representing the forecast error, The probability density function corresponding to the random variable representing the prediction error, The threshold constraint representing the upper bound of the prediction error; The lower boundary of the prediction error under the preset confidence for: ； In the formula, represents the probability that the prediction error exceeds the preset threshold, The threshold constraint representing the lower bound of the prediction error; The probability box-CVaR interval is: ； In the formula, represents the upper boundary of the probability box-CVaR interval, Represents the lower boundary of the probability box-CVaR interval; The final uncertainty fluctuation range satisfy: 。 4. The optimization scheduling method based on probability box and conditional risk value according to claim 1, It is characterized in that The constructing of the optimization scheduling model based on the final uncertainty factor fluctuation range includes: Constructing an objective function and constraints according to the fluctuation range of the final uncertain factors, wherein the constraints include power flow constraints of the distribution network, line transmission capacity constraints, node voltage constraints, and controllable distributed power supply operation constraints; An optimized scheduling model is obtained according to the objective function and the constraint conditions.

5. An optimization scheduling terminal based on probability box and conditional risk value, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, each step of the optimization scheduling method based on probability box and conditional risk value described in any one of claims 1 to 4 is implemented.

Images