数据库读写分离

close all
clear,clc
% 统计数据 实际值
arr = [32709 37165 44021 49125 54203 62380 70323 78375 86349 97345 108625 123653 137804 156728]’;
[m,nn]=size(arr);
alpha = 0.95; % 平滑常数的范围为[0,1]
% 1次指数平滑
s1 = zeros(m,1);
s1(1,1) = arr(1,1);
for i=2:m
s1(i) = alpha*arr(i,1)+(1-alpha)*s1(i-1);
end
sx1 = s1
% 2次指数平滑
s2 = zeros(m,1);
s2(1,1) = arr(1,1);
for i=2:m
s2(i) = alpha*s1(i,1)+(1-alpha)*s2(i-1);
end
sx2 = s2
% 3次指数平滑
s3 = zeros(m,1);
s3(1,1) = arr(1,1);
for i=2:m
s3(i) = alpha*s2(i,1)+(1-alpha)*s3(i-1);
end
sx3 = s3
% 计算二次曲线中的参数
a = zeros(m,1);
b = zeros(m,1);
c = zeros(m,1);
beta=alpha/(2*(1-alpha)*(1-alpha));
a = 3*sx1-3*sx2+sx3;
b = beta*((6-5*alpha)sx1-2(5-4*alpha)*sx2+(4-3*alpha)*sx3);
c = beta*alpha*(sx1-2*sx2+sx3);
% 二次曲线模型 a+b*t+c*t*t
t = 1; %
sf = zeros(m,1);
% sf(1,1) = arr(1,1);
sf = a+b*t+c*t*t % 预测
first = 0;
for i=1:m
first = first+(arr(i,1)-sx1(i,1))*(arr(i,1)-sx1(i,1));
end
second = 0;
for i=1:m
second = second+(arr(i,1)-sx2(i,1))*(arr(i,1)-sx2(i,1));
end
third = 0;
for i=1:m
third = third+(arr(i,1)-sx3(i,1))*(arr(i,1)-sx3(i,1));
end
wrong = zeros(1,3);

wrong(i,1) = first;
wrong(i,2) = second;
wrong(i,3) = third;
test = wrong
asd=a