数据是1978-2017年的GDP，用曲线来拟合序列趋势并对下一年进行预测。代码为：
gdp<-scan("D:/国民生产总值1978-2017.txt")
year<-c(1978:2017)
gdp_data<-data.frame(year,gdp)
gdp_ts<-ts(gdp_data$gdp,start=1978)
t<-c(1:40)
t_2<-t*t
T<-data.frame(gdp,t,t_2)
gdp_fit<-lm(T$gdp~.,data=T)
summary(gdp_fit)
#预测
gdp_fore<-forecast(gdp_fit,h=1,0.95)
gdp_fore
plot(gdp_fore,main="1978-2017年国内生产总值",xlab="年份",ylab="GDP")
但是结果显示为
Error in model.frame.default(Terms, newdata, na.action = na.action, xlev = object$xlevels) :
  变数的长度不一样('t_2')
此外: Warning messages:
1: In forecast.lm(gdp_fit, h = 1, 0.95) :
  newdata column names not specified, defaulting to first variable required.
2: In forecast.lm(gdp_fit, h = 1, 0.95) :
  Could not find required variable t_2 in newdata. Specify newdata as a named data.frame
3: 'newdata' had 1 row but variables found have 40 rows
请教各位大佬是哪里出了问题吗？望赐教，十分感谢！

数据如下
3678.7 4100.5 4587.6 4933.7 5380.5 6043.8 7314.2 9123.6 10375.4 12166.6 15174.4 17188.4 18923.3 22050.3 27208.2 35599.2 48548.2 60356.6 70779.6 78802.9 83817.6 89366.5 99066.1 109276.2 120480.4 136576.3 161415.4 185998.9 219028.5 270844.0 321500.5 348498.5 411265.2 484753.2 539116.5 590422.4 644791.1 686449.6 740598.7 824828.4