CN111800209B - Solar energy prediction method based on energy model and dynamic weight factor - Google Patents

Abstract

The invention relates to a solar energy prediction method based on an energy model and a dynamic weight factor, belonging to the field of wireless communication energy collection. Find the most similar energy model from the historical energy model, and based on this, carry out the energy prediction of the next time slot. The most similar historical energy model is obtained according to the minimum average error of the energy values of the first K time slots of the n+1th time slot of the current day and the energy values of the corresponding time slots in the previous D days. In order to dynamically reflect the influence of weather changes on the prediction results, the present invention sets a dynamic weighting factor, so that the weighting factor can be adjusted accordingly with the weather changes, and can better reflect the contribution of each component in the prediction model to the prediction result, thereby Improve forecast accuracy for solar energy harvesting. The invention is simple, efficient, low in complexity, easy to implement on an actual wireless sensor node, and has good practicability.

Description

A Solar Energy Prediction Method Based on Energy Model and Dynamic Weighting Factors

technical field

The invention belongs to the field of wireless communication energy collection, and relates to a solar energy prediction method based on an energy model and a dynamic weight factor.

Background technique

Wireless sensor networks are widely used in industrial and agricultural production, medical services, environmental monitoring, home security, military and other fields. The continuous energy supply of sensor nodes is a major bottleneck restricting the practical application of wireless sensor networks. Collecting solar energy to supply energy to sensor nodes is an effective way to solve the continuous energy supply of wireless sensor networks. Because solar energy collection is greatly affected by factors such as daytime, weather, and region, it cannot provide a continuous and stable power supply. Therefore, in order to rationally utilize solar energy, it is necessary to predict and manage solar energy collection in order to effectively utilize solar energy. Solar energy prediction is the premise of solar energy management and distribution. It can accurately and effectively predict the short-term and long-term collected solar energy, and provide a basis for subsequent energy distribution and management, so as to effectively improve the utilization efficiency of solar energy. At present, the commonly used solar energy prediction methods mainly include EWMA, WCMA, Pro-Energy and UD-WCMA.

Most of the above methods use the energy value collected in the previous time slot as the basic component of the energy prediction of the next time slot, but in the case of large weather fluctuations, the data of the previous time slot is far from being able to provide a strong force for the next time slot. For reference, this will lead to a sharp increase in prediction error and a greatly reduced algorithm accuracy. At the same time, the weighting factor in most forecasting methods is a fixed value. With the continuous change of weather conditions, the fixed weighting factor cannot follow the changes of the weather, and cannot well reflect the contribution of each component in the forecasting model to the forecasting result. will lead to a decrease in prediction accuracy.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to provide a solar energy prediction method based on an energy model and a dynamic weight factor. This method selects the most similar historical energy harvesting model as the basis of the prediction model, combines the weather change factor, and uses the dynamic weight factor to measure the contribution of the above two parts in real time, which improves the accuracy of energy prediction when the weather changes drastically.

To achieve the above object, the present invention provides the following technical solutions:

A solar energy prediction method based on an energy model and a dynamic weight factor, the method includes the following steps:

S1: Divide a natural day into N time slots, record the solar energy value collected in each time slot, and store the N energy sample values obtained in a natural day as a weather model; record the past D days For the collected solar energy, a total of D historical energy models are obtained;

S2: The current day is recorded as the d day. According to the solar energy values collected in the first K time slots of the n+1th time slot of the day, select a most similar historical energy model E(i ^* ) from the past D days ;

S3: Calculate the weather condition factor GAP _n+1 and the dynamic weight factor α _n+1 of the n+1th time slot;

S4: According to the most similar historical energy model, weather condition factor and dynamic weight factor, the following solar energy prediction method is obtained:

为预测的当天第n+1个时隙的能量值，E(i^*,n+1)为最相似能量模型中第n+1时隙的能量值，GAP_n+1为第n+1个时隙的天气条件因子，M_D(d,n+1)为前D天第n+1时隙的平均能量值。in,

is the predicted energy value of the n+1th time slot of the day, E(i ^* ,n+1) is the energy value of the n+1th time slot in the most similar energy model, and GAP _n+1 is the n+1th time slot The weather condition factor of the time slot, M _D (d,n+1) is the average energy value of the n+1th time slot in the previous D days.

Optionally, in the S2, the specific steps of selecting the most similar historical energy model are as follows:

Calculate the average error of the energy collected by the first K time slots of the n+1th time slot of the current day and the K corresponding time slots of the ith day in the past D days, and expressed as follows:

Among them, E(d,k) and E(i,k) represent the energy collected in the kth time slot on the current day and the ith day, respectively, and p _k-n+K is the corresponding weight; according to the past K time slots The time interval to the current time slot is assigned weights, and p _k-n+K forms a weight vector P, which is expressed as:

The energy model with the smallest average error is the most similar energy model, and the corresponding days are represented by i*; the selection method of the most similar historical energy model in the n+1th time slot of the day is expressed as:

Optionally, in the S3, the specific steps of the weather condition factor GAP are as follows:

The average energy value collected in the n+1th time slot in the past D days is:

Define a vector V=[v ₁ ,v ₂ ,...,v _k-n+K ,...,v _K ] with K elements, and its element v _k-n+K represents the n+1th time of the day The ratio of the energy value of the first K time slots of the slot to the average energy of the corresponding time slots in the past D days, expressed as:

Then the weather condition factor of the n+1th time slot of the day is expressed as:

Optionally, in the S3, the specific steps for calculating the dynamic weight factor α _n+1 are as follows:

Calculate the absolute value λ ₁ of the difference between the product GAP _n ×MD ( _d ,n) and the energy value E(d,n) collected in the nth time slot:

λ ₁ =|GAP _n ×MD ( _d ,n)-E(d,n)| (8)

Calculate the absolute value λ ₂ of the difference between the energy value of the nth time slot in the most similar weather model and the energy value of the nth time slot of the current day:

Then the dynamic weight factor of the n+1th time slot is expressed as:

The beneficial effects of the present invention are:

(1) The present invention is aimed at the problem that the solar energy value collected in the previous time slot cannot provide strong support for the energy prediction of the next time slot when the weather fluctuates greatly, and uses the historical energy model to find the most similar energy model. , based on this, the energy prediction of the next time slot is performed, thereby reducing the prediction error and improving the accuracy of the algorithm.

(2) In the weight setting of the present invention, the weight factor is dynamically calculated for each time slot, so that the weight factor can be adjusted according to the weather changes, which can better reflect the contribution of each component in the prediction model to the prediction result, thereby Improve prediction accuracy.

(3) The present invention makes full use of the historical energy model and dynamically adjusts the weight factor, so that the prediction model can follow the changes of the weather as much as possible, thereby effectively improving the accuracy of solar energy collection prediction; and the proposed method is simple to implement, low in complexity, and easy to use in It is implemented on the actual wireless sensor node and has good practicability.

Other advantages, objects, and features of the present invention will be set forth in the description that follows, and will be apparent to those skilled in the art based on a study of the following, to the extent that is taught in the practice of the present invention. The objectives and other advantages of the present invention may be realized and attained by the following description.

Description of drawings

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be preferably described in detail below with reference to the accompanying drawings, wherein:

Fig. 1 is the time slot allocation diagram of each day;

Figure 2 is a flow chart of the present invention.

Detailed ways

The embodiments of the present invention are described below through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments are only used to illustrate the basic idea of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

Among them, the accompanying drawings are only used for exemplary description, and represent only schematic diagrams, not physical drawings, and should not be construed as limitations of the present invention; in order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings will be omitted, The enlargement or reduction does not represent the size of the actual product; it is understandable to those skilled in the art that some well-known structures and their descriptions in the accompanying drawings may be omitted.

The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms “upper”, “lower”, “left” and “right” , "front", "rear" and other indicated orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the indicated device or element must be It has a specific orientation, is constructed and operated in a specific orientation, so the terms describing the positional relationship in the accompanying drawings are only used for exemplary illustration, and should not be construed as a limitation of the present invention. situation to understand the specific meaning of the above terms.

As shown in FIG. 1 , it is a time slot allocation diagram of the present invention. Divide each day into N time slots, marked as 1, 2, ..., N, and each time slot records the energy value collected by the time slot once. The N energy sample values obtained in one day are stored as a weather model; the solar energy collected in the past D days is recorded, and a total of D historical energy models are obtained.

As shown in FIG. 2 , it is a flow chart of a solar energy prediction method based on an energy model and a dynamic weight factor according to the present invention. Divide a day into N time slots, record the solar energy value collected in each time slot, and store the N energy sample values obtained in one day as a weather model; record the solar energy collected in the past D days , a total of D historical energy models are obtained. The current day is recorded as the d day (d≥D), according to the solar energy values collected in the first K time slots (K≤n) of the n+1th time slot of the day, select a most similar history from the past D days The energy model calculates the weather condition factor and dynamic weighting factor for the n+1th time slot. The predicted energy value of solar energy in the next time slot is obtained from the most similar historical energy model, weather condition factor and dynamic weight factor.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be Modifications or equivalent replacements, without departing from the spirit and scope of the technical solution, should all be included in the scope of the claims of the present invention.

Claims (3)

1. A solar energy prediction method based on an energy model and a dynamic weight factor is characterized in that: the method comprises the following steps:

s1: dividing a natural day into N time slots, recording the solar energy value collected by each time slot, and storing N energy sample values obtained in the natural day as a weather model; recording the solar energy collected in the past D days to obtain D historical energy models;

s2: marking the current day as D day, and selecting a most similar historical energy model E (i) from the past D days according to the solar energy values collected from the first K time slots of the (n +1) th time slot of the current day^*)；

S3: calculating the weather condition factor GAP of the (n +1) th time slot_n+1And a dynamic weight factor alpha_n+1；

S4: according to the most similar historical energy model, the weather condition factor and the dynamic weight factor, the following solar energy prediction method is obtained:

wherein,

for the predicted energy value of the (n +1) th slot of the day, E (i)^*N +1) is the energy value of the (n +1) th time slot in the most similar energy model, GAP_n+1Weather condition factor for the (n +1) th time slot, M_D(D, n +1) is the average energy value of the n +1 time slot of the previous D days;

in S3, a dynamic weighting factor α is calculated_n+1The method comprises the following specific steps:

calculating the product GAP of the weather condition factor and the energy mean value of the nth time slot_n×M_DAbsolute value lambda of the difference between (d, n) and the energy value E (d, n) acquired in the nth time slot₁：

λ₁＝|GAP_n×M_D(d,n)-E(d,n)| (8)

Calculating the absolute difference between the energy value of the nth time slot in the most similar weather model and the energy value of the nth time slot in the current dayFor the value lambda₂：

The dynamic weight factor for the (n +1) th slot is expressed as:

2. the solar energy prediction method based on the energy model and the dynamic weight factor as claimed in claim 1, wherein: in S2, the specific steps of selecting the most similar historical energy model are as follows:

calculating the average error of the energy collected by the first K time slots of the (n +1) th time slot on the day and the K corresponding time slots on the ith day in the past D days, and expressing the average error as follows:

where E (d, k) and E (i, k) represent the energy collected at the k-th slot on the current day and the i-th day, respectively, p_k-n+KIs the corresponding weight; assigning weights, p, according to the time interval from the past K time slots to the current time slot_k-n+KA weight vector P is composed, represented as:

the energy model with the minimum average error is the most similar energy model, and the corresponding daily times are represented by i; the selection method of the most similar historical energy model of the (n +1) th time slot in the current day is represented as follows:

3. the solar energy prediction method based on the energy model and the dynamic weight factor as claimed in claim 1, wherein: in S3, the specific steps of the weather condition factor GAP are as follows:

the average energy values collected at the n +1 th time slot on the past D days are:

defining a vector having K elements, V ═ V₁,v₂,…,v_k-n+K,…,v_K]Element v thereof_k-n+KThe ratio of the energy value of the first K time slots of the (n +1) th time slot of the day to the average energy of the corresponding time slots of the last D days is expressed as:

then the weather condition factor for the (n +1) th slot of the day is expressed as:

Images